6414 Langmuir 2004, 20, 6414-6420

Seeded High Yield Synthesis of Short Au Nanorods in

Aqueous Solution
Tapan K. Sau and Catherine J. Murphy*
Department of Chemistry and Biochemistry, University of South Carolina,
631 Sumter Street, Columbia, South Carolina 29208

Received March 1, 2004. In Final Form: May 11, 2004

Short gold nanorods of average lengths ranging between 20 and 100 nm (with corresponding aspect
ratios of 2 and 4) were synthesized in excellent yield (∼97%). These nanorods were characterized by
dark-field microscopy, UV-visible spectrophotometry, and transmission electron microscopy. Temporal
evolution of rod shape had also been followed by UV-visible spectrophotometry and transmission electron
microscopy and indicates that the nanorods briefly increase in length, then increase slightly in width, as
they grow. The effect of the synthetic parameters on the rod dimension and yield was explored to find out
suitable conditions to produce short nanorods; short nanorods have both plasmon bands in the visible
region of the spectrum, which is a valuable property for sensor applications.

Introduction phineoxide,6 oleic acid,7 and so forth, had been successfully

used for the creation of rod-shaped nanoparticles. Short
Synthesis of nanostructures via simple wet-chemical aspect ratio Au nanorods are especially interesting because
methods is one of the favored routes toward the cost- of their optical properties: they exhibit the transverse as
effective large-scale production of nano-building blocks. well as intense longitudinal plasmon bands in the visible
However, achieving control over the growth of nanostruc- region of the spectrum, making them promising candidates
tures leading to proper dimensional confinement via wet- for sensing and imaging applications.8 In addition, metallic
chemical methods is a challenging task. Of late, there has nanorods that are 100 nm × 200 nm are being explored
been substantial progress in controlling the size and shape for gene delivery applications.9
of nanoparticles by restricted growth obtained by the We have reported2c,4a,b,10 the synthesis of gold and silver
introduction of templates,1 surface capping agents,2 and nanorods/wires with relatively high aspect ratios by
other physicochemical means.3 Solution phase preparation aqueous seeded and nonseeded growth methods. The
of metallic nanorods and nanowires is a challenging task synthesis relies on the reduction of metal salt by a weak
for two reasons. First, surface energy favors the formation reducing agent in the presence of preformed metallic seed
of spherical particles. Second, most metals crystallize in particles. On the basis of electron diffraction analysis and
highly symmetric cubic lattices. Therefore, soft and rigid high-resolution transmission electron microscopy (TEM),
templates1 had been used to achieve rod-shaped metal we have proposed11 a mechanism in which gold nanorods
nanostructures. However, surface capping agents, such evolve by the symmetry breaking of face-centered cubic
as cetyltrimethylammonium bromide (CTAB),4 benzyldi- metallic structures by preferential adsorption of capping
methylhexadecylammonium chloride,5 tetraoctylphos- agents to different crystal faces to produce anisotropic
penta-twinned particles. Recently, Gai and Harmer12 and
Xia et al.13 studied similar systems and gathered further
* Corresponding author. E-mail: murphy@mail.chem.sc.edu.
(1) (a) Pileni, M. P.; Ninham, B. W.; Gulik-Krzywicki, T.; Tanori, J.; evidence supporting this mechanism of rod evolution. To
Lisiecki, I.; Filankembo, A. Adv. Mater. 1999, 11, 1358. (b) Qi, L. M.; make gold nanorods, we have shown2c that the use of silver
Ma, J. M.; Cheng, H. M.; Zhao, Z. G. J. Phys. Chem. B 1997, 101, 3460. nitrate in the seeded synthesis method improved the yield
(c) Li, M.; Schnablegger, M. H.; Mann, S. Nature 1999, 402, 393. (d) as well as subtly changed the shape of gold nanorods.
Sloan, J.; Wright, D. M.; Woo, H. G.; Bailey, S.; Brown, G.; York, A. P.
E.; Coleman, K. S.; Hutchison, J. L.; Green, M. L. H. Chem. Commun. Similar observations were previously made in the elec-
1999, 699. (e) Kyotani, T.; Tsai, L. F.; Tomita, A. Chem. Commun. 1997, trochemical synthesis of Au rods.4c Nikoobakht and El-
701. (f) Martin, B. R.; Dermody, D. J.; Reiss, B. D.; Fang, M. M.; Lyon,
L. A.; Natan, M. J.; Mallouk, T. E. Adv. Mater. 1999, 11, 1021. (g) van (6) (a) Peng, Z. A.; Peng, X. G. J. Am. Chem. Soc. 2002, 124, 3343.
der Zande, B. M. I.; Bohmer, M. R.; Fokkink, L. G. J.; Schonenberger, (b) Peng, X. G.; Manna, L.; Yang, W. D.; Wickham, J.; Scher, E.;
C. Langmuir 2000, 16, 451. (h) Cepak, V. M.; Martin, C. R. J. Phys. Kadavanich, A.; Alivisatos, A. P. Nature 2000, 404, 59.
Chem. B 1998, 102, 9985. (i) Thurn-Albrecht, T.; Schotter, J.; Kastle, (7) Puntes, V. F.; Krishnan, K. M.; Alivisatos, A. P. Science 2001,
G. A.; Emley, N.; Shibauchi, T.; Krusin-Elbaum, L.; Guarini, K.; Black, 291, 2115.
C. T.; Tuominen, M. T.; Russel, T. P. Science 2000, 290, 2126. (8) (a) Haynes, C. L.; Van Duyne, R. P. J Phys. Chem. B 2001, 105,
(2) (a) Manna, L.; Milliron, D. J.; Meisel, A.; Scher, E. C.; Alivisatos, 5599. (b) Sun, Y.; Xia, Y. Analyst 2003, 128, 686. (c) Sönnichsen, C.;
A. P. Nat. Mater. 2003, 2, 382. (b) Sun, Y.; Xia, Y. Science 2002, 298, Franzl, T.; Wilk, T.; von Plessen, G.; Feldmann, J.; Wilson, O.; Mulvaney,
2176. (c) Jana, N. R.; Gearheart, L.; Murphy, C. J. Adv. Mater. 2001, P. Phys. Rev. Lett. 2002, 88, 077402.
13, 1389. (d) Bradley, J. S.; Tesche, B.; Busser, W.; Maase, M.; Reetz, (9) Salem, A. K.; Searson, P. C.; Leong, K. W. Nat. Mater. 2003, 2,
M. T. J. Am. Chem. Soc. 2000, 122, 4631. (e) Antonietti, M.; Grohn, F.; 668.
Hartman, J.; Bronstein, L. Angew. Chem., Int. Ed. Engl. 1997, 36, 2080. (10) (a) Caswell, K. K.; Bender, C. M.; Murphy, C. J. Nano Lett. 2003,
(f) Ahmadi, T. S.; Wang, Z. L.; Green, T. C.; Henglein, A.; El-Sayed, M. 3, 667. (b) Jana, N. R.; Gearheart, L.; Murphy, C. J. Chem. Commun.
A. Science 1996, 272, 1924. 2001, 617. (c) Gao, J.; Bender, C. M.; Murphy, C. J. Langmuir 2003,
(3) Jin, R.; Cao, Y. C.; Hao, E.; Metraux, G. S.; Schatz, G. C.; Mirkin, 19, 9065. (d) Busbee, B. D.; Obare, S. O.; Murphy, C. J. Adv. Mater.
C. A. Nature 2003, 425, 487. 2003, 15, 414.
(4) (a) Murphy, C. J.; Jana, N. R. Adv. Mater. 2002, 14, 80. (b) Jana, (11) Johnson, C. J.; Dujardin, E.; Davis, S. A.; Murphy, C. J.; Mann,
N. R.; Gearheart, L.; Murphy, C. J. J. Phys. Chem. B 2001, 105, 4065. S. J. Mater. Chem. 2002, 12, 1765.
(c) Ying, Y.; Chang, S. S.; Lee, C. L.; Wang, C. R. C. J. Phys. Chem. B (12) Gai, P. L.; Harmer, M. A. Nano Lett. 2002, 2, 771.
1997, 101, 6661. (13) Sun, Y.; Mayers, B.; Herricks, T.; Xia, Y. Nano Lett. 2003, 3,
(5) Nikoobakht, B.; El-Sayed, M. A. Chem. Mater. 2003, 15, 1957. 955.

10.1021/la049463z CCC: $27.50 © 2004 American Chemical Society

Published on Web 06/19/2004
Seeded High Yield Synthesis of Short Au Nanorods Langmuir, Vol. 20, No. 15, 2004 6415

Table 1. Produced Gold Nanorod Dimensions and Yield, with Corresponding Initial Concentrations of Reactantsa
product reaction conditions
dimensionb (length × width) yieldb [Au3+], M [Ag+], M [AA], M [Au]seed, M figure number
87 ((17) × 42 ((10) 97 ((3) 4.0 × 10-4 6.0 × 10-5 6.4 × 10-4 1.25 × 10-7 2a
64 ((12) × 24 ((6) 97 ((3) 4.0 × 10-4 6.0 × 10-5 6.4 × 10-4 2.5 × 10-7 2b
62 ((10) × 23 ((3) 93 ((5) 4.0 × 10-4 6.0 × 10-5 6.4 × 10-4 5.0 × 10-7 2c
50 ((5) × 15 ((3) 90 ((5) 4.0 × 10-4 6.0 × 10-5 6.4 × 10-4 1.25 × 10-6 2d
475 ((24) × 15 ((2) 55 ((12) 4.0 × 10-4 6.4 × 10-4 5.0 × 10-7 2e
80 ((15) × 40 ((10) 97 ((3) 6.0 × 10-4 6.0 × 10-5 9.6 × 10-4 5.0 × 10-7 6a
54 ((10) × 14 ((3) 90 ((5) 3.0 × 10-4 6.0 × 10-5 2.4 × 10-4 5.0 × 10-7 6b
22 ((3) × 6 ((2) 88 ((5) 1.0 × 10-4 6.0 × 10-5 1.6 × 10-4 5.0 × 10-7 6c
90 ((11) × 15 ((2) 57 ((14) 4.0 × 10-4 3.0 × 10-3 5.0 × 10-7 7a
75 ((8) × 10 ((2) 55 ((14) 4.0 × 10-4 3.0 × 10-3 1.5 × 10-6 7b
50 ((6) × 10 ((2) 55 ((17) 4.0 × 10-4 3.0 × 10-3 2.5 × 10-6 7c
a Reactants were added in the order indicated, from left to right. All reactions were run in 5 mL of aqueous 9.5 × 10-2 M CTAB solutions

at room temperature. b A total of 600 particles from three identical batches (200 particles from each) were counted to calculate the rod
yield, and 150 nanorods (50 nanorods from each identical batch) were considered to calculate the average rod dimension. Rod yield is given
by (number of rods)/(total number of particles) × 100%. The error bars in the dimensions correspond to one standard deviation in each
case.

Sayed5 obtained Au rods of varied aspect ratio in solution the reaction mixture was gently mixed for 10 s and left
by varying the amount of silver nitrate for a given amount undisturbed for at least 3 h.
of gold. They reported a very high yield of Au nanorods Characterization. Absorption spectra of the solutions were
when they used surfactant-stabilized Au seeds. Here, we taken on a CARY 500 Scan UV-vis-NIR spectrophotometer.
extended the above methods to produce a very high yield Dark-field microscopy images were taken in a Nikon dark-field
of short Au nanorods of average lengths ranging from 20 microscope system with an oil immersion objective. TEM images
to 100 nm in a controllable manner. We carried out a were obtained either with a Hitachi H-8000 or a JEOL JEM-
systematic study to observe the influence of various 100CX II electron microscope. The TEM grids were prepared as
reaction parameters on the rod dimension. We have shown follows: Typically 1.5 mL of the solution was centrifuged for 12
that in addition to the concentration ratio of Au seed to min at a speed of 14 000 rpm to precipitate the solid. The colorless
Au3+ ions, the reducing agent (L-ascorbic acid, AA) and supernatant was discarded. The solid residue was redispersed
in 1.5 mL of DI water and centrifuged again. Finally the solid
capping agent played important roles in controlling the
residue was dispersed in a suitable volume of DI water depending
aspect ratio as well as the yield of Au nanorods. We also
on the quantity of the residue. A total of 7 µL of this solution was
report on the light-scattering properties of both short and dropcast on the TEM grid and allowed to dry in the open
long nanorods, as observed by dark-field microscopy. atmosphere. Elemental analysis was carried out on a Hitachi
2500 Delta scanning electron microscope by X-ray energy
Experimental Section dispersive analysis (EDAX).
Chemicals. HAuCl4‚3H2O (99.9%), NaBH4 (99%), AA (99+%),
CTAB (99%), and AgNO3 (99+%) were used as purchased
(Aldrich). Ultrapure deionized water (DI; Continental Water
Results and Discussion
Systems) was used for all solution preparations and experiments.
Short gold nanorods are of interest for optical sensing
Glassware was cleaned by soaking in aqua regia and finally
washing with DI water. applications because both plasmon bands are in the visible
Methods. Preparation of Au Seeds. In a typical procedure, region of the spectrum. Our previous syntheses of short
0.250 mL of an aqueous 0.01 M solution of HAuCl4‚3H2O was gold nanorods (aspect ratio less than 6) produced large
added to 7.5 mL of a 0.10 M CTAB solution in a test tube (glass amounts of spherical side products that can be difficult
or plastic). The solutions were gently mixed by the inversion. to separate. The present study was undertaken to find
The solution appeared bright brown-yellow in color. Then, 0.600 the appropriate reaction conditions to prepare short gold
mL of an aqueous 0.01 M ice-cold NaBH4 solution was added all
at once, followed by rapid inversion mixing for 2 min. Care should nanorods, in high yield, by room-temperature colloid
be taken to allow the escape of the evolved gas during mixing. chemical methods.
The solution developed a pale brown-yellow color. Then the test The procedure we have used here for making gold
tube was kept in a water bath maintained at 25 °C for future use. nanorods has the same ingredients as others;2c,4b,5,10d
This seed solution was used 2 h after its preparation and could however, we demonstrate here how the rod length and
be used over a period of 1 week. This preparation differs from
other seed preparations we have used, in that the CTAB width change with time and with variations in concentra-
surfactant is present at this stage. We prepared gold seeds at tions of the reducing agent and stabilizing surfactant, in
four different CTAB concentrations, namely, 9.5 × 10-2, 7.5 × addition to [seed]/[Au3+] ratio. Depending on the reaction
10-2, 5.0 × 10-2, and 8.0 × 10-3 M. Seeds prepared at 8.0 × 10-3 conditions, the rod formation continues for 1 h or so as
M CTAB produced non-rod-shaped and phi-shaped rod particles. judged spectrophotometrically (Figure 1). The longitudinal
Preparation of Au Nanorods. Appropriate quantities of CTAB plasmon band begins to appear in 1 or 2 min, and
solution, water, HAuCl4, AgNO3, AA, and seed solutions were
taken one by one in the order given (see Table 1) in a test tube interestingly, it blue-shifts as the rods develop with time.
and mixed gently by inversion. When the seed solution was added The nanorods at various stages of their growth process
before AA, the reaction became very slow. Therefore, the seed were also characterized by TEM to obtain a vivid picture
solution was always added after the addition of AA. For example, of their evolution (Supporting Information Figure 1). The
in a typical experiment, 4.75 mL of 0.10 M CTAB, 0.200 mL of trend in the change in longitudinal plasmon peak position
0.01 M HAuCl4‚3H2O, and 0.030 mL of 0.01 M AgNO3 solutions with time seems to be in agreement with the change in
were added in that order, one by one, to a test tube, followed by
gentle mixing by inversion. The solution at this stage appeared
aspect ratio (as opposed to length alone) of the developing
bright brown-yellow in color. Then 0.032 mL of 0.10 M AA was rods (Figure 1); the aspect ratio of the rods increases
added to it. The solution became colorless upon addition and quickly, then slowly decreases over time. Schatz et al.
mixing of AA. Finally, 0.010 mL of seed solution was added, and have described a theoretical framework that correlates
6416 Langmuir, Vol. 20, No. 15, 2004 Sau and Murphy

of non-rod-shaped particles were produced (Figure 2;

compare part c with part e, and Table 1, compare the
third row with the fifth). The role of silver nitrate is not
clearly understood at this moment. We proposed2c that
AgNO3 forms AgBr in the presence of CTAB and AgBr
adsorbs differentially to the facets of Au particles, thereby
restricting their growth to the rod shape. Nikoobakht and
El-Sayed support the notion of AgBr on the Au rod surface
and proposed that AgBr decreases the charge density and,
hence, repulsion between the neighboring headgroups
resulting in CTAB template elongation; CTAB in this
interpretation is thought to function as a soft template (it
makes the aspect ratio ∼4 micelles in water).5 In our
experiments, the final nanorod samples contained at most
∼4% silver, according to EDAX; but this measurement
does not distinguish between silver that is alloyed within
gold nanorods and that merely adsorbed to the surface in
either elemental or compound form.
It is believed that well-organized CTAB molecules not
only help to maintain one-dimensional growth but also
dictate the approach rate of gold ions and AA to the seed
particles. There is evidence that CTAB forms bilayers on
gold nanorods and that these bilayers may help “zip” along
the long axis of the nanorods to promote a rodlike
shape.10c,15 In fact, the kinetics of reduction of gold ions
to atomic gold (as followed by the increase in absorbance
at 440 nm)16 showed that the reduction was slower in the
presence of silver nitrate (Figure 3). Furthermore, we
observed that a decrease or an increase in the amount of
silver nitrate added could lead to the formation of non-
rod-shaped (quasi spherical or irregularly grown spiked)
particles. Therefore, the proper kinetics of the growth step
too played a role in the formation of rods in high yield, as
was observed by Peng et al. in the case of CdSe nano-
crystals.6,17
In the case of gold nanorods made without silver ion as
an additive, the nanorods are obtained in relatively low
yield (many spheres are present) but are quite long, with
aspect ratios of ∼20. In this method, in the presence of
silver ion, the gold nanorod yield is nearly quantitative,
but the highest aspect ratio obtainable is ∼5. Under-
standing the mechanism at this point is complicated by
the preliminary result that these short rods, made with
silver ion, appear to have a different crystallography than
the ones made without silver ion.11 In our previous work
without silver ion, the gold nanorods were penta-
tetrahedral twins, in which five {111} triangular facets
were on the ends of the rods, and {100} facets were on the
long sides of the rods.11 In the present case, according to
very preliminary high-resolution TEM images, these short
gold nanorods made in the presence of silver ions are single
crystalline, with possibly {111} facets on the long side of
Figure 1. (a) Visible-NIR absorption spectra of a representa- the rods. Silver bromide is the likely form of silver during
tive gold nanorod solution growing from gold seeds as a function the synthesis (as a result of the ∼0.1 M bromide present
of time from 1 to 60 min after addition of seed to the growth in the seed stage from CTAB; and silver bromide is not
solution. (b) Variation of longitudinal plasmon band maximum
with time. Conditions: [CTAB] ) 9.5 × 10-2 M, [HAuCl4] ) 4.0 reduced to silver metal by ascorbate). Silver bromide can
× 10-4 M, [AgNO3] ) 6.0 × 10-5 M, [AA] ) 6.4 × 10-4 M, and also form thin films epitaxially on Au{111} that can
[Au]seed ) 5.0 × 10-7 M. (c) Temporal evolution of nanorod length, undergo reconstruction.18 Also, bromide ions make incom-
width, and aspect ratio based on TEM micrographs (Supporting mensurate adlayers on both Au{111} and Au{100} that
Information). reconstruct depending on surface coverage.19 Taken
together, should our preliminary crystallography result
plasmon band position with nanoparticle shape, in terms stand, we speculate that bromide adsorbs to the gold seeds;
of aspect ratio, that coincides with our data.14 the presence of silver might form silver bromide at the
Silver nitrate was essential for the preparation of short
Au nanorods in very high yield. For example, when silver (15) Nikoobakht, B.; El-Sayed, M. A. Langmuir 2001, 17, 6368.
nitrate was not added, longer rods with a large fraction (16) Rao, P.; Doremus, R. J. Non-Cryst. Solids 1996, 203, 202.
(17) Peng, X. G. Adv. Mater. 2003, 15, 459.
(18) Mason, M. G.; Hansen, J. C. J. Vac. Sci. Technol., A 1994, 12,
(14) Kelly, K. L.; Coronado, E.; Zhao, L. L.; Schatz, G. C. J. Phys. 2023.
Chem. B 2003, 107, 668. (19) Magnussen, O. M. Chem. Rev. 2002, 102, 679.
Seeded High Yield Synthesis of Short Au Nanorods Langmuir, Vol. 20, No. 15, 2004 6417

Figure 2. Transmission electron micrographs of dense ensembles of gold nanorods synthesized with different reaction conditions,
as a function of seed concentration. [Au]seed were (a) 1.25 × 10-7 M, (b) 2.5 × 10-7 M, (c) 5.0 × 10-7 M, and (d) 1.25 × 10-6 M. The
Au seed concentration in part e was the same as in part c. In all cases, [CTAB] ) 9.5 × 10-2 M, [HAuCl4] ) 4.0 × 10-4 M, [AgNO3]
) 6.0 × 10-5 M, and [AA] ) 6.4 × 10-4 M, except no silver nitrate was used in part e.
gold surface and slow the subsequent gold growth step morphology results from a silver bromide layer on the
(which we do observe compared to the situation in the {111} faces of the gold nanocrystal, leading to gold
absence of silver); and the slower kinetics results in single reduction on other faces to produce a rod with {111} facets
crystalline growth of the gold nanorods and the rod on its long sides.
6418 Langmuir, Vol. 20, No. 15, 2004 Sau and Murphy

ions per seed particle was available for the growth.4a,b,10

The transmission electron micrographs in Figure 2 show
the effects of the gold seed concentration and silver nitrate
on the rod dimension and yield. As observed previously,
the use of a small quantity of silver nitrate had a profound
effect on the rod yield (compared to spheres) and in
controlling rod dimension. Selection of a proper ratio of
seed to gold ion concentrations and addition of an
appropriate quantity of AgNO3 resulted in a very high
yield (∼97%) of rod-shaped particles. Visible-NIR ab-
sorption spectra (Figure 5) too showed that the rod yield
in the absence of AgNO3 was considerably smaller as
judged by the relative values of the longitudinal plasmon
absorbance. Figure 5 shows that in the presence of AgNO3
there was a red shift of the longitudinal plasmon absor-
bance peak position upon an increase in seed concentra-
tion. Furthermore, in addition to the usual transverse
and longitudinal plasmon absorbance peaks, one ad-
Figure 3. Comparison of rates of formation of atomic gold, as ditional medium energy peak could be observed in the
followed by the increase in absorbance at 440 nm in the presence absorption spectra; depending on the faceting of the ends
(a) and absence (b) of silver nitrate. Conditions: in all cases, of the rods, additional plasmon bands can be observed.14
[CTAB] ) 9.5 × 10-2 M, [HAuCl4] ) 4.0 × 10-4 M, [AA] ) 6.4 The red shift with the decrease in rod length appeared to
× 10-4 M, and [Au]seed ) 5.0 × 10-7 M. [AgNO3] ) 6.0 × 10-5 be contradictory but was consistent with their increasing
M in curve a, and no AgNO3 was used in curve b.
aspect ratio, as a result of the simultaneous decrease in
rod width (Table 1).14
Short gold nanorods possess physicochemical properties
that are distinctly different from the longer ones. For Effect of Au3+ Ion Concentration. It had been noted
example, to see the difference in the optical properties in the previous section that the rod dimension strongly
between short Au nanorods and long ones, we compared depends on the total amount of gold ions present in the
the true color of Rayleigh scattered white light from these solution. The variation in rod dimension and yield due to
samples obtained under dark-field conditions in an optical the variation in Au3+ ion concentration was again observed
microscope (Figure 4). As expected, the longer rods to be sensitively dependent on both [Au]seed and [CTAB].
preferentially scatter lower-energy orange/red light For example, for a [Au]seed ) 5 × 10-7 M and [CTAB] )
whereas shorter rods scatter blue-green and yellow light. 9.5 × 10-2 M, we varied the Au3+ ion concentration between
Also, because both plasmon bands are in the visible for 1.0 × 10-4 and 8.0 × 10-4 M, keeping always [AA] ) 1.6-
the short rods but only the transverse one is in the visible [Au3+], and observed that well-defined rod shapes could
for the long rods, the short rods scatter more visible light be obtained up to [Au3+] ∼ 6.0 × 10-4 M. The representative
and appear brighter than the long rods (Figure 4). This TEM images given in Figure 6 demonstrate the effect of
phenomenon can be exploited for optical sensing and [Au3+] variation on the rod dimension. The ends of the
imaging, and studies in this regard, an “optical biochip”, rods appeared more rounded with the decrease in [Au3+].
are in progress. Other workers have used dark-field Visible-NIR absorption spectra of the samples showed
microscopy to image metallic nanoparticles of various that in this case too the shift in the lowest energy plasmon
shapes and have observed that shifts in the color of the absorption peak position was in agreement with the
scattered light are correlated with the adsorbate’s refrac- variation in the aspect ratio of the rods.
tive index; this color change can be used for sensing.20 Effect of AA Concentration. The rod length can also
For the remainder of the paper, we discuss the influence be varied by variation in the AA concentration. The rod
of the reaction parameters one by one for the synthesis length decreases with an increase in [AA] keeping all other
of short gold nanorods. conditions the same. Thus, by increasing the [AA] and
Effect of Seed Concentration. The rod length de- manipulating the [Au]seed/[Au3+] ratio one can form short
creased, and ultimately very short rods with a small Au nanorods in good yield. Figure 7 shows transmission
number of spheres appeared, with an increase in the seed electron micrographs of some short gold nanorod samples
concentration, for a given concentration of gold(III) ion. prepared with a large excess of AA. Depending on the
This trend was expected because a lesser quantity of Au3+ other factors, the rod length could decrease by one-half to

Figure 4. True color images of samples of (a) shorter and (b) longer gold nanorods. The magnification is 100×. The rod dimensions
are length/nanometer × width/nanometer ) 57 ((11) × 19 ((3) and 475 ((24) × 15 ((2) for shorter and longer rods, respectively.
Seeded High Yield Synthesis of Short Au Nanorods Langmuir, Vol. 20, No. 15, 2004 6419

stoichiometric amount of AA required for the reduction of

Au3+ ions to Au0). A comparison of the TEM image in
Figure 7a with that in Figure 1e corroborates this fact.
Peng et al. observed in the case of the CdSe system6,15
that there was a greater tendency to form spherical
particles under the condition of faster supply of monomer
growth units to the seeds. The decrease in rod yield in our
present case also seemed to be due to the very fast
formation and supply of growth units (Au0) to the seeds
in the presence of a large excess of reducing agent (AA),
thereby inducing the growth of seed particles in all
directions and forming more spherical particles. It is not
clear why hardly any rods were obtained under such a
large excess of AA, when silver nitrate was added to the
system. Perhaps silver ions were reduced to atomic silver
(although others have argued against this)5 and, hence,
Figure 5. Visible-NIR absorption spectra of gold nanorod were not available for the participation in template
samples prepared under various reaction conditions, as a elongation via the decrease in headgroup charge of CTAB
function of Au seed concentration. Conditions are as given in molecules by forming AgBr.
Figure 2, spectra a-e corresponding to samples a-e.
Effect of CTAB Concentration. Rod length decreases,
one-fifth for an increase in [AA] from 1.6 to 7.5 times the whereas the width of the rods increases slightly, upon
[Au3+] (i.e., an increase from small excess to 5 times the decreasing CTAB concentration in the absence of AgNO3.

Figure 6. Transmission electron micrographs of gold nanorods synthesized with different amounts of gold salt for a constant seed
concentration. In all cases, [CTAB] ) 9.5 × 10-2 M, [AgNO3] ) 6.0 × 10-5 M, [AA] ) 1.6[HAuCl4], and [Au]seed ) 5.0 × 10-7 M.
[HAuCl4] were 6.0 × 10-4 M in part a, 3.0 × 10-4 M in part b, and 1.0 × 10-4 M in part c.
6420 Langmuir, Vol. 20, No. 15, 2004 Sau and Murphy

Figure 7. Transmission electron micrographs of short gold nanorods prepared in the absence of silver nitrate. In all cases, [CTAB]
) 9.5 × 10-2 M, [HAuCl4] ) 4.0 × 10-4 M, and [AA] ) 3.0 × 10-3 M. [Au]seed )5.0 × 10-7 M in part a, 1.5 × 10-6 M in part b, and
2.5 × 10-6 M in part c.

Unfortunately, the yield of rods drops significantly upon be obtained in high yield. Temporal evolution of gold
decreasing CTAB concentration and, thus, this route nanorods had been followed spectrophotometrically as well
becomes unfavorable for the synthesis of short rods in as by TEM, which revealed how the length and width of
good yield.21 Furthermore, lower CTAB concentrations the developing rods changed with time. Interestingly, the
can lead to non-rod-shaped particles in the presence of nanorods appear to get longer briefly, then “fill out” and
AgNO3 or at higher AA concentrations. get wider, during the course of the reaction. For the
maximum yield of short gold nanorods, silver ion is critical,
Conclusion as is the appropriate concentrations of all other reagents.
In conclusion, we have systematically varied the In the absence of silver ion, longer gold nanorods can be
synthetic parameters involved in the Au rod formation obtained, but significant amounts of spherical side prod-
event and demonstrated how short Au nanorods ranging ucts are produced.
in length from 20 to 100 nm (aspect ratio from 2 to 4) could
Supporting Information Available: Transmission
(20) (a) Mock, J. J.; Smith, D. R.; Schultz, S. Nano Lett. 2003, 3, 485.
(b) Raschke, G.; Kowarik, S.; Franzl, T.; Sonnichsen, C.; Klar, T. A.; electron micrographs of gold nanorods at various stages of their
Feldmann, J.; Nichtl, A.; Kurzinger, K. Nano Lett. 2003, 3, 935. (c) growth. This material is available free of charge via the Internet
McFarland, A. D.; Van Duyne, R. P. Nano Lett. 2003, 3, 1057. at http://pubs.acs.org.
(21) Sau, T. K.; Murphy, C. J. Mater. Res. Soc. Symp. Proc. 2004,
789, 203-212. LA049463Z