In situ synthesis of scalable metallic nanodots in electron microscope

Article abstract of DOI:10.1149/1.3262585

Electron-beam (E-beam) induced reduction has emerged as a technology for the rapid preparation of scalable metallic nanodots. Under a high energy electron bombardment in scanning electron microscopy or transmission electron microscopy, various metallic ions from an aqueous solution are reduced and self-aggregated into metallic nanodots on different host materials. The self-aggregated Ag nanodots prepared at an E-beam voltage of 80 kV exhibit variances in size deviation (σ_D / D) of 13.6% and in displacement deviation (σ_X / X) of 60%. By greatly increasing the E-beam voltage from 3 to 120 kV, the size deviation of the Ag nanodots increases from 1.7 to 7.3 nm, which is mainly attributed to the damage effect and to regrowth at the high accelerating voltage bombardment. The reduction of Cu and Fe nanodots is also demonstrated at an E-beam voltage of 80 kV, indicating a σ_D / D of 11.7% and a σ_X / X of 58.7%. The metallic nanodot diameter of 16±1.9 nm on the carbon nanowires is smaller than that on the carbon fiber (30±3.1 nm), as compared to other geometric host matrices. Furthermore, the results reveal that the nanodot size is precisely confined by decreasing the geometric scale of the host material to a submicrometer range.

Full text of DOI:10.1149/1.3262585

Journal of The Electrochemical Society, 157 ͑2͒ E13-E18 ͑2010͒
E13
0013-4651/2009/157͑2͒/E13/6/$28.00 © The Electrochemical Society
In Situ Synthesis of Scalable Metallic Nanodots in Electron
Microscope
Yi-Hao Pai and Gong-Ru Linz
Graduate Institute of Photonics and Optoelectronics and Department of Electrical Engineering, National
Taiwan University, Taipei 10617, Taiwan
Electron-beam ͑E-beam͒ induced reduction has emerged as a technology for the rapid preparation of scalable metallic nanodots.
Under a high energy electron bombardment in scanning electron microscopy or transmission electron microscopy, various metallic
ions from an aqueous solution are reduced and self-aggregated into metallic nanodots on different host materials. The self-
¯
aggregated Ag nanodots prepared at an E-beam voltage of 80 kV exhibit variances in size deviation ͑␴D/D͒ of 13.6% and in
¯
displacement deviation ͑␴X/X͒ of 60%. By greatly increasing the E-beam voltage from 3 to 120 kV, the size deviation of the Ag
nanodots increases from 1.7 to 7.3 nm, which is mainly attributed to the damage effect and to regrowth at the high accelerating
voltage bombardment. The reduction of Cu and Fe nanodots is also demonstrated at an E-beam voltage of 80 kV, indicating a
¯
¯
␴D/D of 11.7% and a ␴X/X of 58.7%. The metallic nanodot diameter of 16 Ϯ 1.9 nm on the carbon nanowires is smaller than that
on the carbon fiber ͑30 Ϯ 3.1 nm͒, as compared to other geometric host matrices. Furthermore, the results reveal that the nanodot
size is precisely confined by decreasing the geometric scale of the host material to a submicrometer range.
© 2009 The Electrochemical Society. ͓DOI: 10.1149/1.3262585͔ All rights reserved.
Manuscript submitted January 15, 2009; revised manuscript received October 19, 2009. Published December 7, 2009.
Nanoscale metallic dots have emerged as promising nano-objects
and have found potential applications in nanocomposites and elec-
tronic devices. Typically, “top-down” synthesis processes such as
high speed wet milling, the laser ablation technique, and physical
vapor or sputtering-induced cutting have been the focus of compre-
hensive studies.1-3 Alternatively, chemical processes, including oxi-
dation reduction, hydrothermal reduction, and sol–gel synthesis,4-6
as well as other possible reactions, such as pulsed photoirradiation,
high temperature reduction, hydrogen reduction, electrochemical re-
duction, reverse micelles, or the rapid supercritical fluid expansion
of different aqueous solutions, have also been addressed.7-14 Al-
though several technologies have been proposed for preparing nano-
scale metallic dots with controllable uniform size distribution, it has
sometimes been difficult to obtain nanoscale metallic dots of uni-
form sizes in a specific manufacturing system with a strict param-
eter. Therefore, the precise tuning of the processing parameter for
controlling the uniformity of the nanoscale metallic dots on different
substrates has become an important issue, and the chemical or elec-
trochemical reduction of metallic ions into nanodots from solutions
has recently been implemented.12,15-18 However, the size of the non-
uniformly distributed particles cannot be severely confined during
deposition.12,16-18 In view of the previously proposed methodolo-
gies, different metallic nanodots have been achieved using a reduc-
ing agent, such as sodium borohydride, or a protecting agent, such
as citric acid.14,18 During the reduction-based preparation, a strictly
controlled nucleation associated with the prevention of particle co-
agulation is essential, and most of the techniques require strict con-
ditions for preserving nanoparticle size, which tend to be long and
tedious.16-18
Since an electron beam ͑E-beam͒ can also be employed in situ to
locally reduce nanoscale metallic dots from their aqueous
solutions,19,20 commercially available E-beam diagnostic instru-
ments, such as transmission electron microscopy ͑TEM͒ or scanning
electron microscopy ͑SEM͒, are potential candidates for the E-beam
reduction of metallic nanodots. In general, the E-beam bombard-
ment of a surface can simultaneously result in two different catego-
ries of damages: ͑i͒ radiolytic damages, such as electrostatic charg-
ing, contamination, and atomic displacement;21 and ͑ii͒ direct
knock-on collisions of electrons with atomic nuclei and radiolytic
atomic displacements mediated by E-beam-induced electronic
excitations.22 The former can be easily avoided by establishing an
optimal electrical contact or by specimen pretreatment, such as
plasma cleaning. According to the literature,23 the theoretical dis-
placement threshold for an impulse perpendicular to the local tan-
gent plane is roughly half the damage threshold for the impulse.
That is, the electron’s kinetic energy must be greater than the atomic
displacement energy of the irradiated species for the nonrelativistic
energies of SEM. Such an in situ E-beam reduction can be applied to
synthesize metallic nanodots on nanofibers or clay for fabricating
nanoelectrodes, nanocatalysts, and antibiotic materials. However,
discussions of the performance of such a simplified process and its
applications for metallic self-aggregation on versatile host materials
have seldom occurred.
This study investigated the controlled reduction and nucleation
of uniformly size-distributed metallic nanodots from a metallic
aqueous solution based on their self-aggregation under a high energy
electron bombardment in SEM or TEM. During the E-beam irradia-
tion within the SEM or TEM chamber, the monodispersed metallic
nanodots were in situ and locally synthesized on several host mate-
rials without the use of a reducing agent. The relationship between
the E-beam energy and the size/displacement distribution of the re-
duced metallic nanodots has been characterized. The effect of the
host matrices with different geometric scales in confining the metal-
lic nanodot size after self-aggregation has also been discussed.
Experimental
The conceptual description on the E-beam reduction-induced
metallic nanodot synthesis is illustrated in Fig. 1. The concentration
of the aqueous solution for these metal salts was set at 0.005 mol/L.
The homemade lacey carbon film-coated TEM copper grid was ei-
ther dipped in or dropped into a metallic aqueous solution of
AgNO3/H2O, CuSO4/H2O, or FeCl2/H2O.24 After being dried in a
vacuum at room temperature, the lacey carbon film-covered copper
grid samples were placed on a specimen holder and inserted into the
SEM or TEM vacuum chamber. The reduction was initiated right
after the accelerating voltage was turned on. The reaction was dy-
namic and continuous during the 10 min SEM or TEM photograph-
ing procedure. The accelerating voltage of the SEM or TEM ranged
between 3 and 120 kV, and the vacuum in the SEM or TEM cham-
ber was controlled at 2 ϫ 10−6 Torr.
There were two possible reactions in our experiment. One was
discriminated adsorption, which would facilitate the metallic nan-
odot reduction on the lacey carbon film from the aqueous solution
after the E-beam reduction, as the surface carbon dangling bonds on
the lacey carbon film have a propensity for catching metallic
ions.25,26 The other possible reaction was the decomposition of the
AgNO3 dot during the E-beam irradiation and its precipitation again
on the matrix by annihilating the dissolution in the vacuum chamber.
However, the lacey film on clay was simultaneously heated under
z E-mail: grlin@ntu.edu.tw
Downloaded on 2015-05-10 to IP 140.115.130.238 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E14
Journal of The Electrochemical Society, 157 ͑2͒ E13-E18 ͑2010͒
Figure 1. ͑Color online͒ Conceptual de-
scription of the E-beam reduction method
for metallic nanodots fabrication.
the E-beam illumination. Because numerous electrons with ex-
tremely high kinetic energies were bombarded and accumulated on
the lacey film, the relaxation of the hot electrons relied on the heat
transfer between the electron and the lacey film. Due to the rela-
tively low melting point of nanoscale Ag as compared to that of the
bulk Ag,27,28 the metallic Ag nanodot could also be produced by
decomposing the tiny AgNO3 dot on the lacey surface heated by
electrons.29
Results and Discussion
E-beam reduction-inducedAg nanodot synthesis on copper grid.—
To confirm the E-beam reduction-induced Ag nanodot synthesis on
the copper grid, we monitored the SEM photographs of the lacey
carbon film-covered copper grids, which were either untreated or
dipped in a AgNO3 aqueous solution at a concentration of 0.005
mol/L, as shown in Fig. 2. The untreated lacey carbon film-covered
copper grid was a reticulate structure providing significant Cu, O,
and C correlated EDXS signal peaks, as shown on the right of Fig.
2a. In contrast, Fig. 2b shows the SEM image of the lacey carbon
film-covered copper grid after it had been dropped into the AgNO3
aqueous solution of 0.005 mol/L concentration and dried in a
vacuum at room temperature. The insets in Fig. 2a and b are the
magnified images taken by SEM and TEM, respectively. In the inset
of Fig. 2a, it can be clearly seen that nothing was adsorbed on the
lacey carbon film-covered copper grid, which was not dipped in the
AgNO3 aqueous solution. In more detail, a TEM photograph of the
sample was also taken after a 3 kV E-beam reduction in SEM, as
shown in the inset of Fig. 2b. The TEM analysis confirmed the
existence of uniformly distributed Ag nanodots 12.5 Ϯ 1.7 nm in
diameter on the lacey carbon film and the contrast between the im-
During the E-beam bombardment, the Mn+ ions instantly ob-
tained numerous electrons and immediately reduced to M0 nanodots
on the copper grid. By taking the E-beam reduction of the Ag nan-
odot as an example, the sample previously dropped in the aqueous
solution containing Ag+ ions was exposed to the E-beam bombard-
ment. With an E-beam current of 10−6 A, the reduction rapidly oc-
curred at an electron flowing rate larger than 6.2 ϫ 1012 s−1 in
vacuum.29 The reactions in the aqueous solution and E-beam cham-
ber are described as
AgNO3͑s͒ → Ag+͑aq͒ + NO3−͑aq͒ ͑dissolution of salt͒ ͓1͔
Ag+͑aq͒ + e− → Ag͑s͒ ͑in SEM or TEM͒
͓2͔
2AgNO3͑s͒→2Ag͑s͒ + 2NO2͑g͒ + O2͑g͒ ͑in SEM or TEM͒
⌬
͓3͔
1.0
Cu
In more detail, both the metallic and nitrate ions can be adsorbed
on the lacey carbon film by carbon ͑C-͒ dangling bonds and van der
Waals force, respectively. However, the nitrate ions are left in the
aqueous solution due to its weak bonding with the lacey carbon film
by van der Waals force.25,26,30,31
(a)
0.8
0.6
This study further investigated the preparation of metallic nan-
odots on two other host materials: carbon fiber and clay platelet. The
carbon fiber was modified with Ar+ plasma to produce the dangling
bonds. The carbon fiber containing the dangling bonds was dipped
in an aqueous solution of 0.005 mol/L FeCl2 and then dried at room
temperature in the vacuum chamber. Finally, the pretreated carbon
fiber was placed on a sample holder and subjected to an E-beam. In
another case, the clay platelet was dipped in the AgNO3 aqueous
solution at a concentration of 0.005 mol/L, in which the Na+ ions on
the clay interlayer were rapidly exchanged by Ag+ ions. Before the
reduction of the metallic nanodots, the carbon fiber and the clay
platelet were either dropped into or dipped in an aqueous solution
for 30 s. The slightly yellow slurry sample was then dried at room
temperature in a vacuum chamber exposed to the E-beam bombard-
ment.
The morphology and microstructure of the metallic nanodots af-
ter the E-beam reduction were analyzed by using a field-emission
SEM ͑JEOL, 6700F͒, an 80 kV TEM ͑Zeiss, 902A͒, and a 120 kV
TEM ͑JEOL, 1200EX͒. The crystalline phase analysis was per-
formed through the comparison on the selected area diffraction
͑SAD͒ patterns of the E-beam produced metallic nanodots with
those of the standards reported in the JCPDS data file. Elementary
identification was performed by using an energy-dispersive X-ray
spectroscope ͑EDXS, HORIBA EMAX͒. The size, uniformity, and
dispersion ratio of the electron microscopy image analyses were
demonstrated by using related softwares ͑Digital Micrograph 3.3.1
and Origin R7.0 SRO͒. The average size, uniformity, and dispersion
of each sample were determined and confirmed after reproducing
these samples for four times.
0.4
0.2
0.0
O
C
0
1
2
3
4
5
KeV
1.0
0.8
0.6
0.4
0.2
0.0
(b)
Cu
Ag
O
C
Ag
Ag
0
1
2
3
4
5
KeV
Figure 2. Left: The SEM photographs of ͑a͒ the nontreated TEM copper grid
coated with the homemade lacey carbon film and ͑b͒ the Ag nanodot coated
copper grid after 3 kV of E-beam reduction, and the inset shows a magnified
image taken by TEM after the reduction. Right: ͑a͒ EDXS of the nontreated
copper grid and ͑b͒ EDXS analysis for the Ag nanodot coated copper grid
after E-beam reduction.
Downloaded on 2015-05-10 to IP 140.115.130.238 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 157 ͑2͒ E13-E18 ͑2010͒
E15
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
(a)45
(a)
40
35
30
25
20
15
10
5
20
40
60
80
100
120
Accelerating Voltage (kV)
(b)22
66
63
60
57
54
51
48
45
20
(b)
18
16
14
12
20
40
60
80
100
120
Accelerating Voltage (kV)
Figure 4. ͑Color online͒ ͑a͒ The accelerating voltage-dependent average size
and density and ͑b͒ the accelerating voltage-dependent coefficients of vari-
ance in size deviation and displacement deviation prepared by E-beam re-
duction.
literature,32-35 the growth orientation in the present work was the
same as for other methods, such as chemical reduction and sputter-
ing. However, the size distribution diverged with an increase in the
accelerating voltage of up to 120 kV.
Figure 3. The bright-field images of the Ag nanodots prepared by E-beam
reduction with accelerating voltages of ͑a͒ 80 and ͑b͒ 120 kV.
To quantify the relationship between the size/distribution charac-
teristics and the E-beam energy during the reduction, the coefficients
¯
¯
of the size deviation ͑␴D/D͒ and displacement deviation ͑␴X/X͒ for
the self-aggregated Ag nanodots are defined according to the statis-
tical analysis, as described by36
age of the adsorbed Ag nanodots and the lacey carbon branch due to
electronic scattering efficiency, as has been clearly shown.29 The
EDXS analysis, as shown in Fig. 2b, revealed the presence of the
as-grown Ag nanodots after the E-beam reduction.
n
1
2
¯ ¯
To analyze the effect of the E-beam energy on the Ag nanodot
size and distribution characteristics during reduction, we further
compared the in situ TEM results of the two samples with E-beam
reduction at the higher accelerating voltages of 80 and 120 kV, as
shown in Fig. 3a and b. The Ag nanodots synthesized at an acceler-
ating voltage of 80 kV were uniform with an average diameter of
38.6 nm. The corresponding SAD pattern for determining the crys-
talline structure of the Ag nanodots is shown in the inset of Fig. 3a.
The in situ high vacuum SAD analysis verified that the crystallinity
of the as-grown Ag nanodots was of a polycrystalline face-centered
cubic ͑fcc͒ structure. None of the AgO related SAD signals were
identified, which again indicated that the E-beam reduced Ag nan-
odot had not been oxidized at the very beginning. Among the X-ray
diffractions caused by other orientations such as ͑200͒, ͑220͒, and
͑311͒, the preferred orientation was along the ͑111͒ direction with a
lattice spacing of d = 2.359 Å. In contrast with the previous
¯
␴ /D =
D
͑Di − D͒ /D ϫ 100%
͓4͔
͓5͔
ͱ
ͱ
͚
n − 1 i=1
n
1
2
¯ ¯
¯
␴ /X =
X
͑Xi − X͒ /X ϫ 100%
͚
n − 1 i=1
¯
where Di and D is the diameter of each Ag nanodot and the average
¯
number, Xi and X denote the displacement between the Ag nanodots,
and ␴ and ␴ denote the standard deviation in size and displace-
ment, respectively. Note that the coefficient of variance in size de-
viation must remain small to keep the Ag nanodot in the same size,
whereas the coefficient of variance in displacement deviation should
be as large as possible to reach uniform distribution on the host
material surface. Figure 4a and b shows the statistical parameters of
D
X
¯
¯
the distributed Ag nanodots with a ␴D/D of 13.63% and a ␴X/X of
Downloaded on 2015-05-10 to IP 140.115.130.238 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E16
Journal of The Electrochemical Society, 157 ͑2͒ E13-E18 ͑2010͒
(a)
(b)
Figure 6. ͑Color online͒ Left: SEM micrographs of Fe nanodots on the
carbon fiber surface. Inset: The carbon fiber before dipped into an aqueous
solution ͑upper͒ and PtCl4 aqueous solution reduced into Pt nanodots on the
CNWs prepared by E-beam reduction. Inset: The CNW synthesis by using
Fe nanodots on a carbon fiber surface ͑lower͒. Right: The conceptual scheme
of the E-beam reduction-induced metallic nanodots on the carbon-based fi-
ber.
5a and b͒. Furthermore, the SAD analysis of the grown Cu and Fe
nanodots identified their polycrystalline fcc structure. Of all the ori-
ented growths at ͑110͒, ͑200͒, ͑211͒, and ͑220͒ for the E-beam re-
duced Fe nanodots, the main growth was in the ͑110͒ direction with
a corresponding lattice spacing of d = 2.02 Å. From the SAD ob-
servation of the weak Fe2O3 signal with orientations of ͑220͒ and
͑311͒, we also found that some Fe2+ inevitably oxidizes to Fe3+. In
fact, the Fe2+ ion was easily transformed into Fe3+ via oxidation
when leaving the FeCl2 aqueous solution and being immediately
exposed to air. With a limited oxygen content, the Fe2+-containing
materials were favored, whereas the formation of ferric
͑Fe3+-containing͒ materials was favored in a high oxygen environ-
ment. The metallic nanodots made from their corresponding solu-
tions were well dispersed in the host material with a narrow size
distribution. Our analyses confirmed the difficulty of providing
uniform-size distribution for the metallic nanodots at higher accel-
erating voltage.
Figure 5. The bright-field images of ͑a͒ Cu and ͑b͒ Fe nanodots demon-
strated by E-beam reduction with an accelerating voltage of 80 kV.
60% at an accelerating voltage of 80 kV. By enhancing the acceler-
ating voltages from 80 to 120 kV, both the size and the standard
deviation of the Ag nanodots were simultaneously enlarged, whereas
the nanodot density greatly decreased. A similar situation was also
found in Ag nanodots grown at an accelerating voltage of as high as
120 kV, which was mainly attributed to both the damage effect29 and
the regrowth of Ag crystal37 occurring at a high accelerating voltage
bombardment. In addition, the maximum transferable energy was
much lower than the atomic displacement energy. The E-beam bom-
bardment also damaged the samples due to the direct knock-on col-
lisions of the electrons. Typically,38 the electrochemical processes
for metallic reduction usually require a reduction potential of 0.1–10
V, whereas the E-beam instrument-based approaches ͑using TEM or
SEM͒ can provide 103–104 folds of reduction energy.
E-beam reduction-induced Fe, Pt, Ag nanodots on carbon-based
fiber and clay.— The SEM micrograph in Fig. 6 indicates that the
Fe nanodots with an average diameter of 32 Ϯ 4.3 nm tightly ad-
hered to the carbon fiber. A comparison was performed on the size
of Fe nanodots reduced on different host materials, such as the lacey
carbon film and carbon fiber. The result revealed that an apparent
change in the Fe nanodot size from 13.2 to 32 nm accompanied the
decrease in the area density from 2.5 ϫ 1010 to 1.2 ϫ 1010 cm−2
.
Such an Fe nanodot also functioned as a catalyst for the synthesis of
carbon nanowires ͑CNWs͒ using a microwave plasma chemical va-
por deposition system.39 The fiberlike CNWs covered with carbon
fiber flocky had an average diameter of 30 Ϯ 3.1 nm, as shown in
the SEM photographs ͑Fig. 6͒. After completing the CNWs/carbon
fiber host materials, a 0.003 mol/L PtCl4 aqueous solution was
dropped onto the surface of the CNWs/carbon fiber host materials.
Again, we utilized the E-beam reduction method to obtain the re-
duced Pt particles with a diameter of 16 Ϯ 1.9 nm on the CNWs/
carbon fiber surface. By comparing the two different geometric scale
Size uniformity of E-beam reduction-induced metallic nanodots.—
We extended the E-beam reduction method to other metallic aque-
ous solutions. Under an 80 kV E-beam bombardment, the bright-
field image also revealed the synthesis of Cu and Fe nanodots with
diameters of 6.1 Ϯ 0.65 and 13.2 Ϯ 1.55 nm, respectively ͑see Fig.
Downloaded on 2015-05-10 to IP 140.115.130.238 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 157 ͑2͒ E13-E18 ͑2010͒
E17
¯
¯
phology of the Ag nanodots with a ␴D/D of 13.63% and a ␴X/X of
60%, at an accelerating voltage of 80 kV is demonstrated. Increasing
the accelerating voltage of electrons during the E-beam reduction
enlarges both the size and size deviation of the Ag nanodots. A
comparison was performed on the size of metallic nanodots reduced
on different host matrixes. By comparing two different geometric
scale host matrixes, it was found that the nanodot size can be strictly
confined by decreasing the geometric scale of the host material to a
submicrometer range. Because TEM and SEM instruments are
readily available, such an E-beam reduction-induced metallic nan-
odot synthesis should gain popularity for a wide variety of applica-
tions. An extension of this technique into developing nanoelec-
trodes, nanocatalyst, nanofibers, and antibiotic materials is actively
pursued.
Acknowledgments
This work was supported in part by the National Science Council
͑NSC͒ of the Republic of China, under grant no. NSC98-2218-E-
002-022 and no. NSC 98-2623-E-002-002-ET.
National Taiwan University assisted in meeting the publication costs of
this article.
References
1. J. Y. Kelly and J. M. DeSimone, J. Am. Chem. Soc., 130, 5438 ͑2008͒.
2. Y. Ishikawa, K. Kawaguchi, Y. Shimizu, T. Sasaki, and N. Koshizaki, Chem. Phys.
Lett., 428, 426 ͑2006͒.
3. L. Qu and L. Dai, J. Am. Chem. Soc., 127, 10806 ͑2005͒.
4. X. S. Wang, H. Wang, N. Coombs, M. A. Winnik, and I. Manners, J. Am. Chem.
Soc., 127, 8924 ͑2005͒.
5. C. Kim, M. Noh, M. Choi, J. Cho, and B. Park, Chem. Mater., 17, 3297 ͑2005͒.
6. W. F. Yan, S. M. Mahurin, Z. W. Pan, S. H. Overbury, and S. Dai, J. Am. Chem.
Soc., 127, 10480 ͑2005͒.
Figure 7. ͑Color online͒ Left: The photographs of the hydrophilic nature of
ionic clay ͑upper͒ and the Ag nanodots on the clay surface prepared by
E-beam reduction. Right: The conceptual scheme of the E-beam reduction-
induced metallic nanodots on the clay surface.
7. M. Rele, S. Kapoor, and T. Mukherjee, Mater. Res. Bull., 37, 2275 ͑2002͒.
8. D. A. Fleming and M. E. Williams, Langmuir, 20, 3021 ͑2004͒.
9. D. D. Evanoff, Jr. and G. Chumanov, J. Phys. Chem. B, 108, 13948 ͑2004͒.
10. M. C. McLeod, R. S. McHenry, E. J. Beckman, and C. B. Roberts, J. Phys. Chem.
B, 107, 2693 ͑2003͒.
11. M. J. Meziani, H. W. Rollins, L. F. Allard, and Y. P. Sun, J. Phys. Chem. B, 106,
11178 ͑2002͒.
12. N. R. Jana, L. Gearheart, and C. J. Murphy, Chem. Mater., 13, 2313 ͑2001͒.
13. T. Maiyalagan, J. Power Sources, 179, 443 ͑2008͒.
carbon-based host matrices, it was found that the diameter of the
metallic nanodots on the CNWs ͑16 Ϯ 1.9 nm͒ was smaller than
those on the carbon fiber host ͑30 Ϯ 3.1 nm͒. The results also re-
vealed that the nanodot size could be strictly confined by decreasing
the geometric scale of the host material to a submicrometer range.
To implement a full application of the E-beam reduction proce-
dures, we further designed a nanostructure consisting of both nano-
spheres and layered silicate plates ͑clay͒. The E-beam reduction us-
ing the SEM approach was able to control the Ag nanodot size on
clay40 because the hydrophilic nature of the ionic clay assisted the
adsorption and dispersion of metallic ions in an aqueous solution
without organic modifiers or dispersants. With E-beam irradiation,
the bound Ag ions reduced back to Ag atoms and overcame the ionic
charge on the clay surface to self-aggregate as Ag nanodots. In gen-
eral, the trend of ionic exchange was ranked according to the re-
placement priority: M3+ Ͼ M2+ Ͼ M+.41 Accordingly, Ag+ ions
could exchange with sodium ions in the interlayer of the synthetic
clay. When reducing from Ag+ to Ag0 by E-beam reduction, Ag
nanodots with a uniform size distribution could be immobilized on
the negative charge site of the tetrahedron surface. Figure 7 illus-
trates an SEM micrograph of the Ag nanodots with a diameter of
12.5 Ϯ 2.0 nm on the clay surface. Compared with common or-
ganic dispersants,42 the clay enabled the immobilization of the Ag
nanodots through a charge attraction as well as interfered with the
Ag nanodots’ self-aggregation.
14. E. Keceli and S. Ozkar, J. Mol. Catal. A: Chem., 286, 87 ͑2008͒.
15. S. W. Boettcher, S. A. Berg, M. Schierhorn, N. C. Strandwitz, M. C. Lonergan, and
G. D. Stucky, Nano Lett., 8, 3404 ͑2008͒.
16. R. M. Penner, J. Phys. Chem. B, 106, 3339 ͑2002͒.
17. R. Stiger, B. Craft, and R. M. Penner, Langmuir, 15, 790 ͑1999͒.
18. B. Yin, H. Ma, S. Wang, and S. Chen, J. Phys. Chem. B, 107, 8898 ͑2003͒.
19. R. Wahl, M. Mertig, J. Raff, S. Selenska-Pobell, and W. Pompe, Adv. Mater., 13,
736 ͑2001͒.
20. Y. H. Pai, H. F. Huang, Y. C. Chang, C. C. Chou, and F. S. Shieu, J. Power Sources,
159, 878 ͑2006͒.
21. I. Jencic and I. M. Robertson, J. Mater. Res., 11, 2152 ͑1996͒.
22. R. F. Egerton, P. Li, and M. Malac, Micron, 35, 399 ͑2004͒.
23. V. H. Crespi, N. G. Chopra, M. L. Cohen, A. Zettl, and S. G. Louie, Phys. Rev. B,
54, 5927 ͑1996͒.
24. C. T. Nottbohm, A. Beyer, A. S. Sologubenko, I. Ennen, A. Hutten, H. Rosner, W.
Eck, J. Mayer, and A. Golzhauser, Ultramicroscopy, 108, 885 ͑2008͒.
25. H. Wei, J. Li, Y. Wang, and E. Wang, Nanotechnology, 18, 175610 ͑2007͒.
26. M. Otero, F. Rozada, A. Morán, L. F. Calvo, and A. I. García, Desalination, 239,
46 ͑2009͒.
27. J. G. Zhu, Y. Zheng, and G. A. Prinz, J. Appl. Phys., 87, 6668 ͑2000͒.
28. W. Rabaud, L. Saminadayar, D. Mailly, K. Hasselbach, A. Benoit, and B. Etienne,
Phys. Rev. Lett., 86, 3124 ͑2001͒.
29. D. B. Williams and C. B. Carter, Transmission Electron Microscopy, Plenum, New
York ͑1996͒.
30. M. F. Abou Taleb, G. A. Mahmoud, S. M. Elsigeny, and E.-S. A. Hegazy, J.
Hazard. Mater., 159, 372 ͑2008͒.
Conclusions
31. S. Hamoudi, R. Saad, and K. Belkacemi, Ind. Eng. Chem. Res., 46, 8806 ͑2007͒.
32. D. Manno, E. Filippo, M. Di Giulio, and A. Serra, J. Non-Cryst. Solids, 354, 5515
͑2008͒.
33. D. K. Božanić, V. Djoković, J. Blanuša, P. S. Nair, M. K. Georges, and T.
Radhakrishnan, Eur. Phys. J. E, 22, 51 ͑2007͒.
34. K. Abe, Y. Harada, and H. Onoda, Appl. Phys. Lett., 71, 2782 ͑1997͒.
35. J. H. Hsieh, C. M. Wang, and C. Li, Surf. Coat. Technol., 200, 3179 ͑2006͒.
36. C. M. Douglas and C. R. George, Applied Statistics and Probability for Engineers,
4th ed., John Wiley & Sons, New York ͑2007͒.
The in situ scalable E-beam reduction-induced metallic nanodots
synthesis was successfully demonstrated via the tuning on acceler-
ating voltage of a commercially available electron microscope. Ver-
satile functional metallic nanodots growth could be controlled in situ
on host materials after the E-beam reduction. The impact of this
work is to synthesize the metallic nanodots with a large dispersion
ratio by using an in situ E-beam self-assembly process. The mor-
Downloaded on 2015-05-10 to IP 140.115.130.238 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

E18
Journal of The Electrochemical Society, 157 ͑2͒ E13-E18 ͑2010͒
37. P. Kern, Y. Muller, J. Patscheider, and J. Michler, J. Phys. Chem. B, 110, 23660
40. Y. H. Pai, J. H. Ke, C. C. Chou, J. J. Lin, J. M. Zen, and F. S. Shieu, J. Power
Sources, 163, 398 ͑2006͒.
41. D. L. Sparks, Environmental Soil Chemistry, Academic, New York ͑2002͒.
42. C. Deng, J. Weng, Q. Y. Cheng, S. B. Zhou, X. Lu, J. X. Wan, S. X. Qu, B. Feng,
and X. H. Li, Curr. Appl. Phys., 7, 679 ͑2007͒.
͑2006͒.
38. V. R. David, Thermodynamics of Materials, Vol. 1, p. 154, John Wiley & Sons,
New York ͑1995͒.
39. C. C. Chen, C. F. Chen, C. H. Hsu, and I. H. Li, Diamond Relat. Mater., 14, 770
͑2005͒.
Downloaded on 2015-05-10 to IP 140.115.130.238 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).