Large-scale synthesis of silver nanocubes: The role of HCl in promoting cube perfection and monodispersity

Article abstract of DOI:10.1002/anie.200462208

(Chemical Equation Presented) One lump or two! Selective etching of twinned seeds, mediated by HCl and the oxygen in air, resulted in high yields of perfect single-crystal silver nanocubes in a range of sizes (30-130 nm). The chloride ion enhances oxidation and prevents aggregation, while the proton decreases the rate of reduction and facilitates etching through the formation of nitric acid.

Full text of DOI:10.1002/anie.200462208

Communications
plasmonics, surface-enhanced Raman scattering (SERS), as
well as chemical and biological sensing.[2] A wealth of
chemical methods have been developed for the synthesis of
silver and gold nanostructures that have well-controlled
shapes, including triangular plates,[3] cubes,[4] belts,[5] wires,[6]
rods,[7] and branched multipods.[8] Most of these methods,
however, still require improvement in terms of yield, purity,
monodispersity, and scale of synthesis before they will find
use in commercial applications. We and other groups have
established the polyol process as a means to synthesize silver
and gold nanoparticles with controllable shapes in relatively
large quantities.[4–6] Furthermore, we discovered that single-
crystal cubes and tetrahedrons of silver with truncated
corners/edges could be prepared in high yields through the
selective etching and dissolution of twinned seeds by chloride
ions and oxygen (from air).[9] Herein, we report another
highly effective mediator—hydrochloric acid—for the pro-
duction of single-crystal nanocubes. The yield, perfection of
the cube, and monodispersity of the sample are all greatly
improved by employing HCl. More importantly, the robust-
ness of this reaction even allows it to be carried out in a
disposable vial submerged in an oil bath and heated on a
hotplate. The dependence of morphology on the concentra-
tion of hydrochloric acid was also investigated, and the
synthesis was successfully scaled up to produce silver nano-
cubes on the gram scale.
Nanocube Synthesis
Large-Scale Synthesis of Silver Nanocubes: The
Role of HCl in Promoting Cube Perfection and
Monodispersity**
In a typical polyol synthesis, silver atoms are obtained by
reducing AgNO3 with ethylene glycol (EG) through reac-
tions (1) and (2).[10]
Sang Hyuk Im, Yun Tack Lee, Benjamin Wiley, and
Younan Xia*
2 HOCH2CH2OH ! 2 CH3CHO þ 2 H2O
ð1Þ
ð2Þ
2 Agþ þ 2 CH3CHO ! CH3CHOꢀOHCCH3 þ 2 Ag þ 2 Hþ
Shape control of metal nanoparticles has received consider-
able attention in recent years because of the strong correla-
tion between the shape and the chemical, physical, electronic,
optical, magnetic, and catalytic properties of a nanoparticle.[1]
Silver and gold, in particular, have been intensively studied
owing to their numerous applications that include surface
Once the concentration of silver atoms has reached the
supersaturation value, they will start to nucleate and grow
into nanoparticles. At the same time, nitric acid generated
in situ activates a backward reaction that dissolves the solid
silver that is initially formed [Eq. (3)].[11] Through the
introduction of HCl, reaction (3) could be driven further to
the right owing to the formation of more HNO3 from HCl and
AgNO3.
[*] Dr. S. H. Im, Y. T. Lee, Prof. Y. Xia
Department of Chemistry
4 HNO3 þ 3 Ag ! 3 AgNO3 þ NO þ 2 H2O
ð3Þ
University of Washington
Seattle, WA 98195 (USA)
Fax: (+1)206-685-8665
E-mail: xia@chem.washington.edu
On the basis of these reactions and experimental obser-
vations, we can explain why and how perfect single-crystal
silver nanocubes were formed in high yields. First, upon
addition of silver nitrate and PVP (poly(vinyl pyrrolidone)) to
the hot solution of EG, both twinned and single-crystal seeds
of silver are formed through homogeneous nucleation, with
the twinned particles being the most abundant morphology as
a result of their relatively lower surface energies.[12] These
initially formed nanoparticles are dissolved owing to the
relatively high concentration of HNO3 present in the early
stages of the reaction. As the reaction continues, HNO3 is
gradually consumed and a second round of nucleation occurs.
At very small sizes, it is expected that the crystal structure of
these nuclei fluctuates, as has been observed in several TEM
B. Wiley
Department of Chemical Engineering
University of Washington
Seattle, WA 98195 (USA)
[**] This work was supported in part by ONR (N-00014-01-1-0976) and a
fellowship from the David and Lucile Packard Foundation. Y.X. is an
Alfred P. Sloan Research Fellow and a Camille Dreyfus Teacher
Scholar. Both S.H.I. and Y.T.L. have been partially supported by the
Postdoctoral Fellowship Program of the Korean Science and
Engineering Foundation (KOSEF). B.W. thanks the Center for
Nanotechnology at the University of Washington for an IGERT
Fellowship Award supported by the NSF (DGE-9987620).
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
2154
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200462208
Angew. Chem. Int. Ed. 2005, 44, 2154 –2157

Angewandte
Chemie
studies.[13] When the nanoparticles grow in dimension, they
will be locked into either a single-crystalline or twinned
morphology. While the twinned particles have a lower overall
surface energy, this comes at the expense of significant lattice
distortion and surface defects. Thus, twinned particles are
expected to exhibit a stronger reactivity and susceptibility
towards etching. As no lattice distortion is required to form a
single crystal, these seeds should be relatively more stable in
this environment and can continue to grow. Consequently,
through the selective etching of twinned seeds by HNO3, high
yields of single-crystal nanocubes result (see Supporting
Information for a schematic of this mechanism).
While selective etching is the dominant mechanism for
producing nanocubes, there are additional elements that
contribute to its success. In the present study, both the proton
and the chloride ion of HCl play a significant role. As well as
its function in increasing the concentration of HNO3, the
proton can greatly reduce the net reaction rate according to
Le Chatelierꢀs principle. At the same time, chloride ions likely
adsorb onto the surfaces of silver seeds and thereby prevent
agglomeration through electrostatic stabilization.[14] We also
lowered the reaction temperature from 160 to 1408C to
further slow down the net reaction rate in an effort to increase
the efficiency of selective etching by nitric acid. This
combination of factors allies to promote the production of
silver nanocubes with high yields and monodispersed sizes.
The primary stages of the reaction can be recognized by
their distinctive colors. As illustrated in Figure 1a, the
solution turned milky white after the injection of the solutions
of AgNO3 and PVP. This color suggests the presence of AgCl
precipitate caused by the relatively high concentration of
chloride ions in the reaction mixture (0.25 mm). Figure 1e
shows a TEM image of a sample taken from the suspension at
this stage. The chemical composition of these particles was
AgCl, as confirmed by both X-ray diffraction (XRD) and
energy dispersive X-ray (EDX) analyses. At t = 45 min, the
solution turned light yellow in color (Figure 1b) which
indicates that the AgCl particles had dissolved and only Ag
nanoparticles remained in the reaction mixture. As shown by
TEM (Figure 1 f), Ag nanoparticles of two different sizes
(diameter, d ꢁ 20 and ꢁ 5 nm) coexist in the solution at this
time. The light yellow color gradually faded, and the solution
appeared transparent and colorless at around t = 105 min
(Figure 1c) which suggests the complete dissolution of all
large Ag nanoparticles. This picture is also consistent with
TEM observations (Figure 1g) in which only very few Ag
nanoparticles of very small size (d ꢁ 6 nm) could be found.
This transparent state lasted for approximately one hour,
after which time the solution acquired a reddish tint whose
intensity increased over a period of several hours. By t = 15 h,
the solution had become red (as shown in Figure 1d),
implying the formation of small Ag nanocubes. A typical
TEM image (Figure 1h) of these Ag nanocubes shows that
they were perfect in shape and approximately 30 nm in edge
length. These Ag nanocubes could further grow into cubes of
larger sizes if the reaction was allowed to continue. Finally,
the solution became ocher in color at approximately t = 26 h.
The Ag nanocubes contained in this solution were also perfect
in shape with their edge length increased to around 130 nm.
Figure 1. Morphological revolution with reaction time. a)–d) Photo-
graphs of the reaction solution at t=4, 46, and 103 min, and 15 h,
respectively. e)–h) TEM images of silver nanoparticles contained in the
corresponding reaction solution shown on the left (a–d).
This observation implies that it will be possible to control the
size and therefore the optical properties of Ag nanocubes
simply by varying the reaction time. Figure 2 demonstrates
the size-dependent absorption spectra of the cubes. The
number of peaks and relative positions are consistent with
theoretical calculations.[15]
The dependence of morphology on the concentration of
HCl was also examined. Figure 3a shows the SEM image of a
product obtained at t = 25 h when the concentration of HCl
was 0.125 mm in the final mixture. This sample contained a
mixture of polydisperse silver nanocubes, tetrahedrons, and
Angew. Chem. Int. Ed. 2005, 44, 2154 –2157
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2155

Communications
chloride ions could slow down the etching of twinned seeds by
selectively blocking the twin sites through surface adsorption.
Alternatively, it is possible that an increase in the concen-
tration of HCl may result in the formation of more AgCl
precipitate at the initial stage of the synthesis and that some of
these AgCl colloids may survive and serve as seeds for the
subsequent growth of twinned particles. On the basis of the
results of these experiments, it is concluded that 0.25 mm
represents an ideal concentration of HCl for the synthesis of
silver nanocubes.
To separate the roles played by the proton and chloride,
we also replaced HCl (0.25 mm) with HNO3 (0.25 mm). The
NO3ꢀ ions from HNO3 should have a negligible effect on the
synthesis because the concentration of HNO3 was extremely
low as compared to the concentration of AgNO3. Figure 3d
shows a typical SEM image of the product at t = 30 h that
comprised a mixture of small silver nanocubes (edge length
ꢁ 40 nm) and some irregular particles. This observation
implies that HNO3 alone is able to induce the etching and
dissolution of twinned seeds and thus channel the product
into single-crystal nanocubes. However, owing to the absence
of Clꢀ ions in the solution, the single-crystal seeds could not
be stabilized and therefore might agglomerate into larger,
irregularly shaped particles. It is also expected that a
combination of Clꢀ ions and O2 will facilitate the etching
and dissolution of twinned seeds to further decrease the
percentage of twinned particles in the final product.
Figure 2. UV/Vis absorption spectra of aqueous solutions that contain
silver nanocubes with different edge lengths (30, 45, and 60 nm).
It is an important issue to scale up the synthesis from the
viewpoint of commercial usage. To this end, we increased the
volumes of all solutions by a factor of five. In this case, the
synthesis followed the same pattern of color changes which
suggests that the nucleation and growth mechanisms did not
change as the reaction volume was increased. Figure 4a shows
Figure 3. SEM images of silver nanoparticles synthesized at different
concentrations of HCl (scale bar: 500 nm): a) 0.125, b) 0.25, and
c) 0.375 mm, respectively. The reaction solution contained AgNO3
(23.5 mm) and PVP (36.7 mm, calculated in terms of the repeating
unit) and was heated in an oil bath held at 1408C. d) Silver nanoparti-
cles obtained under the same conditions except that HNO3 (0.25 mm)
was used instead of HCl (0.25 mm).
nanowires. It is believed that the wires and irregular particles
form as a result of the incomplete etching of twinned seeds by
the lower concentration of HNO3. Figure 3b shows an SEM
image of the final product obtained at t = 26 h when the
concentration of HCl was increased to 0.25 mm (namely, the
synthesis described in Figure 1). The solution contained only
monodisperse silver nanocubes of about 130 nm in edge
length. As the concentration of HCl was further increased to
0.375 mm, a different morphology was observed (Figure 3c).
This product, obtained at t = 22 h, was characterized by a
mixture of relatively thick wires and irregular particles. The
mechanism behind this morphological variant is yet to be
understood. In this case, it was found that the reaction
mixture did not become transparent which indicates that the
twinned seeds were not dissolved to cut short their growth
into nanowires and irregular particles. It is assumed that
Figure 4. a) Photograph of the reaction mixture in a scale-up (ꢀ5) syn-
thesis. b) Typical SEM images of the as-synthesized silver nanocubes.
The inset shows a magnified SEM image that illustrates the sharp cor-
ners and edges of these nanocubes. c) An XRD pattern of the same
batch of silver nanocubes. d) TEM image of the silver nanocubes. The
inset shows an electron diffraction pattern recorded by directing the
electron beam perpendicular to the (100) facet of a silver nanocube.
2156
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2005, 44, 2154 –2157

Angewandte
Chemie
spectrometer (Palo Alto, CA) using a quartz cuvette with an optical
path of 1 cm.
a photograph of the final product, which displayed an ocher
color similar to that seen for the smaller reaction volume.
Figure 4b shows a typical SEM image of this sample and
indicates that all particles were cubic in shape with an average
edge length of 125 nm. The inset shows a tilted SEM image at
higher magnification that clearly displays the sharp corners
and edges of these nanocubes. Figure 4c shows an XRD
pattern recorded from the same batch of silver nanocubes.
The abnormal intensity of the (200) peak suggests that the
sample exclusively comprises nanocubes that were preferen-
tially oriented with their (100) planes parallel to the support-
ing substrate. Figure 4d shows a typical TEM image of the
silver nanocubes. Again, it is clear that these silver nanocubes
are single crystals with sharp corners and edges. The inset
shows an electron diffraction pattern recorded by directing
the electron beam perpendicular to the (100) facet of an
individual nanocube and confirms that the particles are single
crystals.
Received: October 5, 2004
Published online: February 28, 2005
Keywords: aggregation · crystal growth · nanostructures ·
.
polyol synthesis · silver
[1] a) Y. Xia, P. Yang, Y. Sun, Y. Wu, B. Gate, Y. Yin, F. Kim, H. Yan,
Adv. Mater. 2003, 15, 353; b) Y. Wu, B. Messer, P. Yang, Adv.
Mater. 2001, 13, 1487; c) N. I. Kovtyukhova, T. E. Mallouk,
Chem. Eur. J. 2002, 8, 4354; d) C. Salzemann, I. Lisiecki, A.
Brioude, J. Urban, M.-P. Pileni, J. Phys. Chem. B 2004, 108,
13243; e) N. R. Jana, L. Gearheart, S. O. Obare, C. J. Murphy,
Langmuir 2002, 18, 922; f) A. T. Ngo, M. P. Pileni, Colloids Surf.
A
2003, 228, 107; g) V. F. Puntes, K. M. Krishnan, A. P.
Alivisatos, Science 2001, 291, 2115; h) Z. L. Wang, T. S.
Ahmad, M. A. El-Sayed, Surf. Sci. 1997, 380, 302.
[2] a) D. Roy, J. Fendler, Adv. Mater. 2004, 16, 479; b) J. R. Krenn,
Nat. Mater. 2003, 2, 210; c) K. Kneipp, Y. Wang, H. Kneipp, L. T.
Perelman, I. Itzkan, R. R. Dasari, M. S. Feld, Phys. Rev. Lett.
1997, 78, 1667; d) J. P. Kottmann, O. J. F. Martin, D. R. Smith, S.
Schultz, Phys. Rev. B 2001, 64, 235402; e) A. J. Haes, R. P.
Van Duyne, J. Am. Chem. Soc. 2002, 124, 10596; f) B. D. Moore,
L. Stevenson, A. Watt, S. Flitsch, N. H. Turner, C. Cassidy, D.
Graham, Nat. Biotechnol. 2004, 22, 1133; g) Y. Cao, R. Jin, C.
Mirkin, Science 2002, 297, 1536.
In summary, monodispersed nanocubes of silver have
been synthesized in large quantities by introducing a small
amount of hydrochloric acid to the conventional polyol
synthesis. From the color changes involved and electron
microscopy studies, it is believed that hydrochloric acid plays
an important role in selectively etching and dissolving
twinned silver nanoparticles. In contrast to our previous
work based on the mediation of NaCl, the current process is
more robust as it involves two different types of etchants:
HNO3 and Clꢀ/O2. Furthermore, the presence of protons
greatly slows down the reduction reaction and thereby
facilitates the formation of single-crystal seeds. Under the
optimized conditions, we have been able to scale up the
synthesis to 0.25 grams (or 1.2 ꢁ 1021 130-nm-wide cubes).
[3] a) R. Jin, Y. Cao, C. A. Mirkin, K. L. Kelly, G. G. Schatz, J. G.
Zheng, Science 2001, 294, 1901.
[4] a) Y. Sun, Y. Xia, Science 2002, 298, 2176; b) R. Jin, S. Egusa,
N. F. Scherer, J. Am. Chem. Soc. 2004, 126, 9900; c) T. K. Sau,
C. J. Murphy, J. Am. Chem. Soc. 2004, 126, 8648; d) F. Kim, S.
Connor, H. Song, T. Kuykendall, P. Yang, Angew. Chem. 2004,
43, 3759; Angew. Chem. Int. Ed. 2004, 43, 3673.
[5] Y. Sun, B. Mayers, Y. Xia, Nano. Lett. 2003, 5, 675.
[6] a) Y. Sun, B. Mayers, T. Herricks, Y. Xia, Nano Lett. 2003, 3, 955;
b) Y. Sun, Y. Yin, B. Mayers, T. Herricks, Y. Xia, Chem. Mater.
2002, 14, 4736; c) Y. Sun, B. Mayers, Y. Xia, Adv. Mater. 2003, 15,
641.
[7] a) N. R. Jana, L. Gearheart, C. J. Murphy, J. Phys. Chem. B 2001,
105, 4065; b) C. J. Murphy, N. R. Jana, Adv. Mater. 2002, 14, 80;
c) F. Kim, J. H. Song, P. Yang, J. Am. Chem. Soc. 2002, 124,
14316.
[8] a) E. Hao, R. C. Bailey, G. C. Schartz, J. T. Hupp, S. Li, Nano
Lett. 2004, 4, 327; b) S. Chen, Z. L. Wang, J. Ballato, S. H.
Foulger, D. L. Carroll, J. Am. Chem. Soc. 2003, 125, 16186.
[9] B. Wiley, T. Herricks, Y. Sun, Y. Xia, Nano Lett. 2004, 4, 1733.
[10] a) C. ꢂzmetin, M. C¸ opur, A. Yartasi, M. M. Kocakerim, Chem.
Eng. Technol. 2003, 23, 707; b) B. Blin, F. Fievet, D. Beaupꢃre, M.
Figlarz, Nouv. J. Chim. 1989, 13, 67.
[11] F. Fievet, J. P. Lagier, M. Figlarz, MRS Bull. 1989, 14, 29.
[12] Z. L. Wang, J. Phys. Chem. B 2000, 104, 1153; b) C. Cleveland, U.
Landman, J. Chem. Phys. 1991, 94, 7376; c) P. M. Ajayan, L. D.
Marks, Phys. Rev. Lett. 1988, 60, 585.
Experimental Section
In a typical synthesis, ethylene glycol (EG; 5 mL, J. T. Baker, 9300-01)
was placed in a 20-mL vial, capped, and heated with stirring in an oil
bath at 1408C for 1 h. HCl (1 mL of a 3 mm solution in EG) was then
quickly added, and the vial was recapped. After 10 min, AgNO3
(3 mL of a 94 mm solution in EG; Aldrich, 209139–100G) and
poly(vinyl pyrrolidone) (PVP; 3 mL of a solution in EG (147 mm in
terms of the repeating unit); Mr ꢁ 55000, Aldrich, 856568-100G) were
simultaneously added with a two-channel syringe pump (KDS-200,
Stoelting, Wood Dale, IL) at a rate of 45 mL per hour to the stirring
solution. The vial was then capped and heated at 1408C. Upon
injection of the solution of AgNO3, the reaction mixture went through
a series of color changes that included milky white, light yellow,
transparent, red, and ocher. To separate the roles of the proton and
chloride, we performed a synthesis under the same conditions except
for the replacement of HCl by HNO3. For the scale-up synthesis, the
vial was replaced with a 100-mL flask, and the volumes of all solutions
were increased by a factor of five.
[13] a) S. Iijima, T. Ichihashi, Phys. Rev. Lett. 1986, 56, 616; b) D. J.
Smith, A. K. Petford Long, L. R. Wallenberg, J. O. Bovin,
Science 1986, 233, 872; c) N. Doraiswamy, L. Marks, Surf. Sci.
1996, 348, L67.
[14] a) C. A. Mirkin, Inorg. Chem. 2000, 39, 2258; b) S. Lin, Y. Tsai,
C. Chen, C. Lin, C. Chen, J. Phys. Chem. B 2004, 108, 2134.
[15] I. O. Sosa, C. Noguez, R. G. Barrera, J. Phys. Chem. B 107, 6269.
All samples for morphology and structure analysis were washed
with acetone and then with water to remove excess EG and PVP.
SEM images were taken using a field emission scanning electron
microscope (FEI, Sirion XL) operated at an accelerating voltage of
10–20 kV. The transmission electron microscopy (TEM) images and
diffraction patterns were obtained using
a JEOL microscope
(1200EX II) operating at 80 kV. X-ray diffraction (XRD) studies
were performed on a Philips 1820 diffractometer with a scan rate of
0.2 degrees per minute in the range 20–908. UV/Vis absorption
spectra were taken at room temperature on a Hewlett Packard 8452
Angew. Chem. Int. Ed. 2005, 44, 2154 –2157
www.angewandte.org
ꢀ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2157