Room temperature synthesis of silver nanowires from tabular silver bromide crystals in the presence of gelatin

Article abstract of DOI:10.1016/j.jssc.2005.11.021

Long silver nanowires were synthesized at room temperature by a simple and fast process derived from the development of photographic films. A film consisting of an emulsion of tabular silver bromide grains in gelatin was treated with a photographic developer (4-(methylamino)phenol sulfate (metol), citric acid) in the presence of additional aqueous silver nitrate. The silver nanowires have lengths of more than 50 μm, some even more than 100 μm, and average diameters of about 80 nm. Approximately, 70% of the metallic silver formed in the reduction consists of silver nanowires. Selected area electron diffraction (SAED) results indicate that the silver nanowires grow along the [111] direction. It was found that the presence of gelatin, tabular silver bromide crystals and silver ions in solution are essential for the formation of the silver nanowires. The nanowires appear to originate from the edges of the silver bromide crystals. They were characterized by transmission electron microscopy (TEM), SAED, scanning electron microscopy (SEM), and powder X-ray diffraction (XRD).

Full text of DOI:10.1016/j.jssc.2005.11.021

ARTICLE IN PRESS
Journal of Solid State Chemistry 179 (2006) 696–701
www.elsevier.com/locate/jssc
Room temperature synthesis of silver nanowires from tabular silver
bromide crystals in the presence of gelatin
a
Suwen Liu , Rudolf J. Wehmschulte , Guoda Lian , Christopher M. Burba
b,
Ã
a
a
a
Department of Chemistry and Biochemistry, University of Oklahoma, 620 Parrignton Oval, Room 208, Norman, OK 73019, USA
b
Department of Chemistry, Florida Institute of Technology, 150 West University Blvd., Melbourne, FL 32901, USA
Received 12 September 2005; received in revised form 10 November 2005; accepted 18 November 2005
Available online 4 January 2006
Abstract
Long silver nanowires were synthesized at room temperature by a simple and fast process derived from the development of
photographic films. A film consisting of an emulsion of tabular silver bromide grains in gelatin was treated with a photographic
developer (4-(methylamino)phenol sulfate (metol), citric acid) in the presence of additional aqueous silver nitrate. The silver nanowires
have lengths of more than 50 mm, some even more than 100 mm, and average diameters of about 80 nm. Approximately, 70% of the
metallic silver formed in the reduction consists of silver nanowires. Selected area electron diffraction (SAED) results indicate that the
silver nanowires grow along the [111] direction. It was found that the presence of gelatin, tabular silver bromide crystals and silver ions in
solution are essential for the formation of the silver nanowires. The nanowires appear to originate from the edges of the silver bromide
crystals. They were characterized by transmission electron microscopy (TEM), SAED, scanning electron microscopy (SEM), and powder
X-ray diffraction (XRD).
r 2005 Elsevier Inc. All rights reserved.
Keywords: Nanowire; Silver; Silver bromide; Tabular; Gelatin; Reduction; Photography; Developer
1
. Introduction
as templates [3,11]. Typically, ‘‘hard’’ templates are made
of solid materials containing channels, pores, or steps [12].
Examples for ‘‘soft’’ templates include DNA chains [13],
peptide nanotubes [14], polyvinylpyrollidone (PVP) [7], or
glycolipid nanotubes [15].
Highly anisotropic noble metal nanowires have attracted
considerable interest due to their potential applications in
optical, electronic, and mechanical nanodevices [1–3].
Silver is especially attractive because it exhibits the highest
electrical and thermal conductivities among all metals, and
it has been used in a wide variety of commercial
applications that range from catalysis and electronics to
photonics and photography [4]. As a result, the synthesis
and characterization of nanostructured silver have recently
attracted the interests of numerous research groups [5–10].
The most widely used methods for the preparation of 1-D
nanomaterials can be loosely summarized as template-
directed syntheses. Generally, a material is deposited into
or onto the template to afford the desired shape followed
by separation of the nanostructured product from the
template. Both ‘‘hard’’ and ‘‘soft’’ materials have been used
The formation of small silver particles by the reduction
of silver salts is the heart of photography [16]. Several years
ago, it was reported that the photographic process could
also be applied for the preparation of silver nanowires [17].
Typically, the active part of a photographic film consists of
a dispersion of silver bromide crystallites in gelatin.
Exposure to light generates silver clusters on the surface
of the silver bromide crystallites, which serve as seeds for
the generation of the large silver particles that make up the
image. Usually, these silver particles are assemblies of silver
nanofilaments with a structure similar to that of a wad of
steel wool [16,18]. It has been shown that variations in the
reaction conditions of the reduction of the silver salts, i.e.
the development, can lead to different silver nanostruc-
tures. For example, thick short silver needles with
diameters around 200 nm and lengths of up to 4 mm were
Ã
Corresponding author. Fax: +1 321 674 8951.
E-mail address: rwehmsch@fit.edu (R.J. Wehmschulte).
0022-4596/$ - see front matter r 2005 Elsevier Inc. All rights reserved.
doi:10.1016/j.jssc.2005.11.021

ARTICLE IN PRESS
S. Liu et al. / Journal of Solid State Chemistry 179 (2006) 696–701
697
obtained from octahedral silver bromide grains [19].
Longer silver nanowires (about 9 mm) were prepared from
silver bromide nanocrystals, albeit in low yields [17].
Application of a diluted developer resulted in the forma-
tion of short thick silver nanofilaments starting from
tabular silver bromide crystallites [20]. Several growth
mechanisms have been proposed for the formation of the
silver filaments, but the details are still not completely
understood [16]. For example, neither the role of the
gelatin nor the shapes and sizes of the silver bromide
crystallites have been included in the proposed mechan-
isms.
refrigerator for storage. Gelatin weight percent: 5.17%,
Gelatin:silver ratio ¼ 1:1.39.
2.3. Film preparation
Glass slides or plastic substrates were coated with
liquefied (45 1C) AgBr/gelatin emulsion. The films that
had been coated onto plastic substrates can be peeled off to
afford freestanding films. All samples were prepared not in
a darkroom but under the fluorescent light in the
laboratory.
Here we report a simple method for the synthesis of
silver nanowires with an average diameter of ca. 80 nm and
lengths of several tens of micrometers, some even over
2
.4. Reducing agent [25]
Solution A: 4-(methylamino)phenol sulfate (metol, 2 g,
.8 mmol) and citric acid (10 g, 4.7 mmol) were dissolved in
1
00 mm, in good yields (over 70% by weight). Contrary to
5
water to give 100 mL solution.
the preparation of photographic films [21] the silver
bromide emulsions used in this study were neither
chemically nor spectrally sensitized, and all experimental
procedures could be performed under fluorescent labora-
tory light rather than in a darkroom. The methodology is
derived from the one reported by Liu et al. [17], but this
contribution focuses on yield and quality improvement and
partial mechanistic understanding.
Solution B: AgNO in water (10% by weight). Immedi-
3
ately before use, two parts of solution B were added to 100
parts of solution A. Dilute reducing agent: 100 mL 1/2
concentration of solution A and 2 mL 1/10 concentration
solution B.
2
.5. Development and fixation
2
. Experimental section
The slides or plastic substrates covered with the AgBr/
gelatin emulsion were submerged in a solution of the
reducing agent, i.e. the developer, for a certain time,
followed by submerging in a fixer, and finally washed by
submerging in distilled water.
2
.1. Starting materials
A cattle bone-type photographic grade gelatin (Bloom
75 g) was a gift from Kodak. Silver nitrate of purity
2
4
99.9% was purchased from Alfa Aesar, potassium
2
.6. Replica
bromide (AR) from Mallinckrodt. The Kodak developer
D-72 and Kodak F-5 fixer were prepared according to
literature procedures [22]. Tabular silver bromide crystals
were obtained according to modified literature procedures
as outlined below. Silver bromide crystal size, shape and
dispersity are strongly dependent on the reaction condi-
tions [21,23,24].
The developed AgBr was separated from the bulk gelatin
by centrifugation of a dilute sample. The AgBr grains and
silver particles or nanowires were then deposited on a cover
glass and dried. Shadowing with tungsten was carried out
at an angle of 451, followed by covering with carbon using
a JEOL JFB 900. Finally, the carbon replicas, with the
silver and silver bromide particles still attached, were
released from the cover glass by treatment with dilute
hydrofluoric acid and washed with water. The silver
bromide was removed by further treatment with sodium
thiosulfate solution. After washing, a copper grid was used
to fish out the floating samples.
2.2. Synthesis of an emulsion of tabular silver bromide and
gelatin
Gelatin (0.75 g) was allowed to swell by soaking in
distilled water (50 mL) for 30 min at 45 1C. KBr (0.59 g,
.0 mmol) was added, and the mixture was stirred at 45 1C
for 50 min. Solutions of AgNO (10 mL, 2 M) and KBr
5
3
(
10 mL, 2 M) were added dropwise to the gelatin solution at
2.7. TEM samples
the same rate and time. The resulting emulsion was stirred
for 15 min and then cooled to 4 1C overnight. The solidified
emulsion was cut into small pieces (0.5 cm in diameter), and
the pieces were washed with distilled water six times each to
After development the slides with the films were treated
with the Kodak F-5 fixer for 4 min. Parts of the films broke
into fragments and floated on top of the fixer solution.
These very thin fragments (approximately 250 nm thick)
were transferred to distilled water and loaded onto the
transmission electron microscopy (TEM) grids after
20 min. Thus, the TEM images show the exact morphology
of the products in the film after reduction.
remove dissolved KBr and KNO (i.e. in each wash cycle
3
the pieces were immersed in distilled water (100 mL) for
15 min). Finally, a gelatin solution (10 mL, 7.5% by weight)
was added to the emulsion, the mixture was homogenized
by stirring at 45 1C for 15 min and cooled to 4 1C in a

ARTICLE IN PRESS
S. Liu et al. / Journal of Solid State Chemistry 179 (2006) 696–701
698
Table 1
Post-fixation results for tabular AgBr samples
Sample/property
Development time
min)
Color of wet sample Color of dry sample Shape
Particle size (nm)
Film no.
(
Ag-48
Ag-49
10
45
Bright yellow
Gray green
Pale purple
Gray
Sphere
Triangle, hexagon
30–50
100
32821–27
32813–20
2
.8. Film sections
The dried freestanding AgBr/gelatin film was embedded
in resin. After the resin solidified, it was cut into sections
with a thickness of approximately 80 nm by microtomy.
These sections were put on a Cu TEM grid.
2
.9. Characterization
TEM was performed on a JEOL TEM-2000 at 200 kV
and scanning electron microscopy (SEM) using a JEOL
80 SEM. X-ray diffraction (XRD) patterns were taken
8
with a SCINTAG XRD Xtra X-ray diffractometer (CuKa,
radiation, 20 kV). The coated slides themselves were used
for XRD measurements.
Fig. 1. (a) SEM image of tabular AgBr crystals. (b) TEM image of the
carbon replica of tabular AgBr crystals after removal of AgBr with the F-5
fixer.
2
.10. Post-fixation process
The exposed silver bromide was dissolved from the slide
before development leaving the silver nanoparticles or
clusters in the gelatin layer. This was achieved by
immersion of the slides containing the tabular AgBr/
gelatin emulsion in 10% Na S O solution for 4 min
2
2
3
followed by washing for three times in distilled water.
The development times were varied, and the results are
shown in Table 1. To obtain samples for TEM the resulting
film was dissolved in 45 1C deionized water. The silver
nanoparticles were separated by centrifugation followed by
re-suspending in warm water. This process was repeated
three times. A few drops of water were added to the silver
deposit to produce an aqueous suspension, which was
dropped onto the copper grids for TEM analysis.
Fig. 2. Silver filaments from development with D-72: (a) 2 min develop-
ment, (b) 4 min development.
Treatment of these emulsions with different developers
under systematically varied reaction conditions afforded
silver nanowires and other nanoparticles. The products
were generally characterized by TEM and in selected
instances also by SEM and powder XRD. For comparison
with commercial films, some of the gelatin emulsions were
treated with a standard commercial developer, D-72. In
this case, tangled and non-uniform nanofilaments were
obtained (Fig. 2), which is typical for the form of silver
found in photographic films after development [16].
Long silver nanowires were obtained in good yields by
application of a mild reducing agent in the presence of
silver nitrate. Fig. 3 shows a typical TEM image of these
nanowires with diameters in the range of 60–80 nm, and
lengths of up to several tens of micrometers with some
wires reaching lengths of more than 100 mm. It should be
3
. Results and discussion
Gelatin emulsions of tabular silver bromide crystals were
prepared according to modified literature procedures
21,23,24]. The SEM image (Fig. 1a) illustrates the typical
[
morphology of the silver bromide crystals, which were
obtained as a mixture of triangular and hexagonal plates
with an average size of about 1 mm (edge length) and a
thickness of about 100–150 nm. The TEM image of a
carbon replica after removal of the silver bromide grains
with sodium thiosulfate (Fig. 1b) shows the presence of
silver nanoparticles that have been on the surface of the
silver bromide crystals. These nanoparticles were generated
by exposure of the emulsion to ambient laboratory light.

ARTICLE IN PRESS
S. Liu et al. / Journal of Solid State Chemistry 179 (2006) 696–701
699
Fig. 3. A typical low magnification TEM image of silver nanowires
obtained from tabular AgBr crystals after 5 min development.
Fig. 4. A close-up image of an individual silver nanowire. The inset shows
an electron diffraction pattern from this wire indicating the growth
direction as [111].
mentioned that these images reflect exactly how the silver
nanostructures existed in the gelatin film, because the TEM
samples consisted of the top layer of the gelatin film.
Although some silver nanoparticles and short filaments
were also found in the products, the nanowires’ weight
percentage was estimated at more than 70%. This is
significantly higher than the 30% reported previously [17].
The TEM image of an individual silver nanowire is shown
in Fig. 4. Several selected area electron diffraction (SAED)
images taken in different areas of the nanowire suggest that
the nanowire is single crystalline cubic silver with a growth
direction along the [111] vector.
The penetration depth of the developer was probed by
TEM on thin cross-sections of free standing developed
gelatin film. Fig. 5 shows that the silver particles and
nanowires are concentrated in the top 250–300 nm layer,
and none were found deeper than 700 nm. Apparently, the
reaction is confined to the surface layer, and this may be
+
explained in two ways: (i) the reducing agents and the Ag
ions diffuse only slowly into the gelatin film during the
5
while permeating the gelatin film.
+
min reaction time or (ii) the Ag ions were consumed
In order to gain deeper understanding of the growth
mechanism of the silver nanowires, a series of experiments
with systematically varied reaction conditions was con-
ducted. Treatment of a gelatin film that did not contain
silver bromide with our developer (solutions A and B) led
to the deposition of metallic silver on the surface of the
gelatin as well as on the wall of the beaker. Only irregularly
shaped silver was formed, and no nanowires were detected.
Treatment of silver bromide grains in the absence of gelatin
under otherwise identical conditions similarly afforded
Fig. 5. TEM image of a sectioned sample of silver nanowires inside the
gelatin film.

ARTICLE IN PRESS
S. Liu et al. / Journal of Solid State Chemistry 179 (2006) 696–701
700
only coarse silver particles. Next, a suspension of silver
nanoparticles in gelatin was generated by exposure of the
silver bromide/gelatin emulsion to laboratory light fol-
lowed by removal of the silver bromide with sodium
thiosulfate solution (post-fixation process). Under these
conditions the silver nanoparticles that have formed on the
surface of the silver bromide grains (see Fig. 1b) remain
suspended within the gelatin. Treatment of this emulsion
with our developer again did not lead to silver nanowires,
but instead to formation of various silver particles
depending on the reaction time. A brief (10 min) contact
with the developer gave oval-shaped silver nanoparticles
with an average size of about 30 nm (Fig. 6a). A longer
reaction time (45 min) afforded well-shaped hexagonal or
truncated trigonal silver prisms with an average size of
were caused by photodecomposition (Fig. 1b). Fig. 7
illustrates the silver nanowires formed at different reducing
times. After a 2 min reduction some nanoparticles and
short nanowires (nanorods) were obtained (Fig. 7a). These
nanorods may be viewed as the initial stage of the nanowire
formation. Increase of the contact time to 5 min (Fig. 3)
resulted in a mixture of nanoparticles and nanowires, with
the nanowires already comprising more than 50% of the
silver. A high ratio of long nanowires to nanoparticles was
observed after 10 min reduction time (Fig. 7b), which
did not change significantly using a 20 min development
(Fig. 7c). The silver nanowires routinely grew to tens of
micrometers in length, some to even more than 100 mm. In
most cases, it can be seen that the nanowires originate from
the edges or sides of the tabular silver bromide grains.
Since the silver bromide crystals remain intact during
development, the silver source is most likely the silver
nitrate solution that is part of the developer, a process that
is known in photography as physical development [16].
The XRD patterns taken from the film samples show
that the silver nanowires and silver nanofilaments synthe-
sized here existed solely in the fcc phase (Fig. 8). The lattice
constant of the silver nanowires calculated from the XRD
1
obtained recently by reduction of AgNO with L-ascorbic
00 nm. These particles may be compared with those
3
acid in the presence of silver seeds and cetyltrimethylam-
monium bromide [26]. Finally, the development times of
the silver bromide/gelatin emulsion were varied, and the
products were analyzed by TEM using carbon replicas. As
silver halides are sensitive to an electron beam, this
technique is commonly used to investigate the morphology
of silver halides [23] as well as the growth mechanism of the
of silver filaments in the photographic process [19]. Silver
nanowires at different stages of development were ob-
served. Before reduction, there were some silver nanopar-
ticles on the surfaces of the silver bromide grains, which
˚
pattern was 4.081 A, which is very close to the reported
˚
data (a ¼ 4:0862 A, JCPDS File 4-0783). It is noteworthy
that the peak width for silver nanowires (Fig. 8b) is slightly
(002)*
(111)
(
(
200)
202)*
(222)*
(220)
(113)
(a)
(400)*
(204)*
(
(
b)
c)
30
35
40
45
50
55
60
65
70
75
80
2
ꢀ degrees
Fig. 8. Powder XRD patterns for (a) silver nanowires before removal of
AgBr, (b) silver nanowires and (c) silver nanofilaments. AgBr reflections
are indexed with*.
Fig. 6. TEM images of post-fixation samples developed for (a) 10 min and
(b) 45 min using diluted developer.
Fig. 7. TEM images of the silver nanowires at different stages of development: (a) after 2 min development, (b) after 10 min development, (c) after 20 min
development.

ARTICLE IN PRESS
S. Liu et al. / Journal of Solid State Chemistry 179 (2006) 696–701
701
narrower than that for silver nanofilaments (Fig. 8a),
indicating a larger average crystallite size of the nanowires
EPSCoR project (EPS-0132354). We also thank G. Strout
for assistance with the TEM.
[
27]. This is consistent with the TEM results, which have
shown that the silver nanowires are thicker than the
nanofilaments.
The mechanism of the silver nanowire formation is not
yet completely understood. However, it has now been
demonstrated that the necessary requirements are at least
References
[
1] Z.L. Wang (Ed.), Nanowires and Nanobelts: Materials, Properties,
and Devices, Kluwer Academic Publishers, Boston, 2003.
2] M.A. El-Sayed, Acc. Chem. Res. 34 (2001) 257–264.
(
i) a film consisting of an emulsion of tabular silver
[
bromide in gelatin, (ii) a mild reducing agent, and (iii) the
presence of silver ions in solutions. Lack of any of these
ingredients does not afford any nanowires. It is interesting
to note that the small silver nanoparticles on the surface of
the silver bromide crystals apparently do not act as seeds
for the silver nanowire growth. On the other hand, it is
common in photographic films that silver nanofilaments
originate at the edges of the silver bromide grains, although
the reason is not clear [20]. The developer is mild enough
not to quickly reduce the crystalline silver bromide but
mainly the silver ions in solution. It has been suggested that
this process may be facilitated by the silver bromide [20].
Silver ions at an edge of a silver bromide grain are reduced,
and the silver ions in the grain are replenished by silver ions
from the solution. This could account for the here observed
nanowire formation. Although the role of gelatin is not yet
understood, it may be speculated that it serves a similar
role as PVP in that it binds preferably to certain crystal
faces [7]. Furthermore, gelatin may also control the
diffusion of developer and silver ions and thus limit the
concentration of silver ions at the growing structure. Silver
has been shown to have an affinity for arginine, cysteine,
lysine and methionine [28], and these are also common in
gelatin [16].
[3] Y. Xia, P. Yang, Y. Sun, Y. Wu, B. Mayers, B. Gates, Y. Yin,
F. Kim, H. Yan, Adv. Mater. 15 (2003) 353–389.
[
4] B. Wiley, Y. Sun, B. Mayers, Y. Xia, Chem. Eur. J. 11 (2005)
54–463.
5] N.R. Jana, L. Gearheart, C.J. Murphy, Chem. Commun. (2001)
17–618.
[6] B.H. Hong, S.C. Bae, C.-W. Lee, S. Jeong, K.S. Kim, Science 294
2001) 348–351.
7] Y. Sun, Y. Yin, B.T. Mayers, T. Herricks, Y. Xia, Chem. Mater. 14
2002) 4736–4745.
8] H. Yan, S.H. Park, G. Finkelstein, J.H. Reif, T.H. LaBean, Science
01 (2003) 1882–1884.
[9] G. Viau, J.-Y. Piquemal, M. Esparrica, D. Ung, N. Chakroune,
F. Warmont, F. Fievet, Chem. Commun. (2003) 2216–2217.
4
[
6
(
[
[
(
3
´
[
[
10] J.-Q. Hu, Q. Chen, Z.-X. Xie, G.-B. Han, R.-H. Wang, B. Ren,
Y. Zhang, Z.-L. Yang, Z.-Q. Tian, Adv. Funct. Mater. 14 (2004)
1
83–189.
11] Y. Sun, Y. Xia, in: Z.L. Wang (Ed.), Nanowires and Nanobelts:
Materials, Properties, and Devices, Kluwer Academic Publishers,
Boston, 2003, pp. 211–234.
[
[
12] M. Barbic, J.J. Mock, D.R. Smith, S. Schultz, J. Appl. Phys. 91
(
2002) 9341–9345.
13] E. Braun, Y. Eichen, U. Sivan, G. Ben-Yoseph, Nature 391 (1998)
75–778.
[14] M. Reches, E. Gazit, Science 300 (2003) 625–627.
7
[
[
[
15] B. Yang, S. Kamiya, Y. Shimizu, N. Koshizaki, T. Shimizu, Chem.
Mater. 16 (2004) 2826–2831.
16] T.H. James (Ed.), The Theory of the Photographic Process,
Macmillan, New York, 1977.
17] S. Liu, J. Yue, A. Gedanken, Adv. Mater. 13 (2001) 656–658.
4
. Summary
[18] W. Van Renterghem, D. Schryvers, J. Van Landuyt, C. Van Roost,
R. De Keyzer, J. Imaging Sci. Technol. 47 (2003) 133–138.
[
[
[
19] R.B. Flint, G.C. Farnell, J. Photogr. Sci. 21 (1973) 175–179.
20] G.C. Farnell, R.B. Flint, J. Photogr. Sci. 29 (1981) 236–242.
21] T. Maternaghan, in: R.B. McKay (Ed.), Technological Applications
of Dispersions, Surfactant Science Series, Marcel Dekker, New York,
1994, pp. 373–441.
A fast and facile procedure for the preparation of long
silver nanowires was found in the reduction of aqueous
silver ions inside a film consisting of a silver bromide/
gelatin emulsion in a procedure similar to the development
of photographic films. The reaction is fast (10–15 min),
proceeds at room temperature on the laboratory desktop
under ambient light and affords straight nanowires in high
yields. Investigations of the mechanism as well as attempts
to increase the yield and selectivity are under way.
[
[
[
[
22] A. Adams, The New Ansel Adams Photography Series, Little Brown,
Boston, 1983.
23] T. Tani, Photographic Sensitivity: Theory and Mechanism, Oxford
University Press, New York, 1995.
24] T.A. Larichev, E.I. Kagakin, Microsc. Res. Technol. 42 (1998)
139–144.
25] J.E. Mack, M.J. Martin, The Photographic Process, McGraw-Hill,
New York, 1939.
Acknowledgments
[
[
26] S. Chen, D.L. Carroll, J. Phys. Chem. B 108 (2004) 5500–5506.
27] C. Hammond, The Basics of Crystallography and Diffraction,
Oxford University Press, New York, 2001.
This work was performed at the University of Oklahoma
and was supported by the Oklahoma NanoNet, an NSF
[28] L.C. Gruen, Biochim. Biophys. Acta 386 (1975) 270–274.