Structural control of Au and Au-Pd nanoparticles by selecting capping ligands with varied electronic and steric effects

Article abstract of DOI:10.1139/V09-127

Weakly interacting ligands including three Gemini surfactants, didodecyldimethylammonium bromide (DDAB), and amines (RNH₂,R ₂NH, and R₃N) were used to prepare Au nanoparticles (NPs). Aqueous Au NPs capped with DDAB and Gemini surfactants showed similar sizes (3-4 nm), whereas toluene-based NPs stabilized with DDAB, amines, and their mixtures range from 2.5 to 9.3 nm. Ligand effect on Au-Pd NP structure was also studied with EXAFS. These findings were consistently accounted for by considering the ligand's electronic/steric effects and mixed ligands coadsorption, and suggest useful ways to control NP structure by manipulating the two effects and using mixed capping ligands.

Full text of DOI:10.1139/V09-127

1641
Structural control of Au and Au–Pd nanoparticles
by selecting capping ligands with varied
electronic and steric effects
Patrick B. Murphy, Feng Liu, Stephen C. Cook, Nusrat Jahan,
D. Gerrard Marangoni, T. Bruce Grindley, and Peng Zhang
Abstract: Weakly interacting ligands including three Gemini surfactants, didodecyldimethylammonium bromide (DDAB),
and amines (RNH2, R2NH, and R3N) were used to prepare Au nanoparticles (NPs). Aqueous Au NPs capped with DDAB
and Gemini surfactants showed similar sizes (3–4 nm), whereas toluene-based NPs stabilized with DDAB, amines, and
their mixtures range from 2.5 to 9.3 nm. Ligand effect on Au–Pd NP structure was also studied with EXAFS. These find-
ings were consistently accounted for by considering the ligand’s electronic/steric effects and mixed ligands coadsorption,
and suggest useful ways to control NP structure by manipulating the two effects and using mixed capping ligands.
Key words: metal nanoparticles, capping ligands, electronic effect, steric effect, EXAFS.
´
´
´
´
Resume : On a utilise des ligands interagissant faiblement, dont trois agents de surface Gemini, le bromure de didodecyl-
´
´
dimethylammonium (BDDA) et des amines (RNH2), R2NH et R3N) pour preparer des nanoparticules (NP) d’or (Au). Les
´
tailles des nanoparticules d’or aqueuses, fermees avec du BDDA ou des agents de surfaces Gemini sont assez semblables
`
`
´
´
(3–4 nm) alors que celles des nanoparticules a base de toluene et stabilisees par du BDDA, des amines ou leurs melanges
`
´
varient de 2,5 a 9,3 nm. On a aussi examine l’effet de ligand sur la structure des nanoparticules Au–Pd par la technique
´
ˆ
´
´
de structure fine d’absorption des rayons-X etendue (SFAXE). Les observations peuvent etre expliquees d’une fac¸on cohe-
´
´
´
`
rente en considerant les effets sterique/electronique du ligand et la coadsorption des ligands mixtes et ceci suggere des fa-
ˆ
c¸ons utiles de controler la structure des nanoparticules en manipulant les deux effets et en utilisant des ligands mixtes pour
les fermer.
´
´
´
´
Mots-cles : nanoparticules metalliques, ligands de fermeture, effet electronique, effet sterique, structure fine d’absorption
´
des rayons X etendue (SFAXE).
´
[Traduit par la Redaction]
acting ligands can offer excellent control of NP structures,
making possible the synthesis of NPs with flexibly con-
trolled size and shape (e.g., nanorods and nanowires).3,11 In
addition, weakly interacting ligands are easily replaceable/
exchangeable if strongly binding ligands are desired to
surface-functionalize the NP surface.
Tetraalkylammonium surfactants are among the most
widely studied weakly interacting ligands in recent years.
Numerous syntheses of NPs have been reported using tetra-
alkylammonium ligands, such as tetraoctylammonium
bromide (TOAB),12 didodecyldimethylammonium bromide
(DDAB),13 and cetyltrimethylammonium bromide (CTAB).11
Remarkably, significant progress has also been made in
understanding the binding/stabilization mechanism of NP
capping ligands, which plays a critically important role in
the development of more efficient NP synthetic routes.2,4,14
However, most of the in-depth studies on the stabilization
mechanisms of weakly interacting ligands are based on ionic
Introduction
Selection of capping ligands for the synthesis of metal
nanoparticles (NPs) is of fundamental and technological im-
portance.1–6 Capping the NP surface with alkanethiols and
their derivatives has become a widely used method in the
past decade.1 The availability of a large variety of commer-
cial thiol derivatives and the advancement in thiol ligand-ex-
change reactions for the NPs have made such an approach
particularly attractive.7–9 However, the primary limitation of
using thiol derivatives to functionalize NP surface is that the
metal–ligand interaction is too strong and this hinders some
applications such as catalysis. An alternative route to control
the NP surface structure is to use weakly interacting mole-
cules (relative to thiols) such as tetraalkylammonium surfac-
tants. A noteworthy advantage of using weakly binding
ligands is that the catalytic activity of NPs can be dramati-
cally improved relative to those of their thiol-stabilized
counterparts.4,10 More significantly, the use of weakly inter-
Received 22 May 2009. Accepted 20 July 2009. Published on the NRC Research Press Web site at canjchem.nrc.ca on 28 October 2009.
P.B. Murphy, F. Liu, S.C. Cook, N. Jahan, T.B. Grindley, and P. Zhang.1 Department of Chemistry and Institute for Research in
Materials, Dalhousie University, Halifax, NS B3H 4J3, Canada.
D.G. Marangoni. Department of Chemistry, St. Francis Xavier University, Antigonish, NS B2G 2W5, Canada.
1Corresponding author (e-mail: peng.zhang@dal.ca).
Can. J. Chem. 87: 1641–1649 (2009)
doi:10.1139/V09-127
Published by NRC Research Press

1642
Can. J. Chem. Vol. 87, 2009
Scheme 1. Molecular structures of three Gemini surfactants, three fatty amines, and DDAB.
ligands.4 For instance, one of the widely accepted ligand-
stabilization mechanisms used for NP catalysis, proposed
by Ott and Finke, mainly consider two effects of the ligands,
i.e., the so-called electronic effect (also known as charge
or DLVO effect) and the steric effect.4 Such a mechanism
has been found to provide good interpretation of the struc-
tural and catalytic properties of high-charge ionic ligands
such as (Bu4N+)9 (P2W15Nb3O62)9–. In contrast, detailed
stabilization mechanisms of other types of ligands, such as
neutral weakly interacting ligands, has remained largely
under-explored. Such studies are particularly desirable in
light of the recent remarkable findings on amine-based nano-
electronic devices.15–17 Herein, we report the synthesis of Au
NPs with a variety of weakly interacting ligands (Scheme 1),
including some novel Gemini surfactants, didodecyldimethy-
lammonium bromide (DDAB), dodecylamine, didecylamine,
tridodecylamine, and nine mixtures of DDAB/amines
(DDAB mixed with three amine ligands, each with three dif-
ferent molar ratios). The purposes of the present work are
threefold: (i) exploring how Finke’s models on ligand’s elec-
tronic and steric effects can be used/extended to understand
the stabilization of Au/Au–Pd NPs with the above-
mentioned ligands (particularly the neutral amine ligands);
(ii) seeking possible ways to controllably synthesize Au NPs
by selecting ligands with varied electronic/steric effects or
using a mixed capping method, and (iii) investigating how
the capping ligands influence the local structure (i.e., the bi-
metal mixing pattern) of Au–Pd NPs.
Experimental section
All the chemicals were obtained commercially from Alfa
Aesar or Aldrich except for ligands Gem A, Gem B, and
Gem C (Scheme 1). Synthetic details for the preparations of
Gemini surfactants will be published elsewhere18 and can
also be found in the Supplementary data. Au nanoparticles
(Au NPs) were prepared by reducing HAuCl4 in aqueous
solution or AuCl3 in toluene with sodium borohydride.
Three amine capping ligands were used singularly or in
combination with DDAB. For combination systems, the ra-
tios of amine to DDAB used were 1:4, 1:1, and 4:1. In a
typical procedure, 0.03 mmol gold(III) chloride and
0.45 mmol of capping ligands (gold:ligand = 1:15 molar ra-
tio) were dispersed in 30 mL of toluene by sonication. Then,
the gold(III) was reduced with sodium borohydride
(0.3 mmol) in 20 mL of water, and stirred under argon for
2 h at room temperature. The as-prepared Au NP liquid sus-
pensions were then stored in darkness in a refrigerator with-
out further purification. UV–vis experiments were conducted
Published by NRC Research Press

Murphy et al.
1643
Scheme 2. A schematic demonstration of the electronic and steric effect of ionic (DDAB) and neutral (R2NH) ligands.
right after the NP synthesis, and TEM measurements were
done within three days. Although only as-prepared samples
(ideally purified samples are preferred) were measured in
the present work, the reported TEM and UV–vis results
were proved to be valid by time-dependent UV–vis meas-
urements. Indeed, the as-prepared Au NPs were found to be
stable for at least one month, as shown by comparing the
UV–vis spectra of fresh and one month old samples. The bi-
metallic Au–Pd NPs capped with DDAB/RNH2 (80%
RNH2) were synthesized using the same conditions as those
reported recently for TOPB/RNH2–capped Au–Pd NPs.19 It
was assumed that the initial Au-to-Pd molar ratio in the
starting materials was maintained after the Au–Pd NPs were
formed, which was supported by the XPS composition anal-
ysis result reported for similar Au–Pd NPs in our recent
publication.19
The morphology of the nanoparticles was characterized
with a Technai 12 or JEOL 123 transmission electron micro-
scope (TEM), both equipped with a CCD camera and oper-
ated at a voltage of 80 keV. One or two drops of the
nanoparticle solutions were cast to carbon or formvar-coated
Cu grids for the TEM measurements. The particle size infor-
mation was obtained using a size-measuring program from
the TEM by counting at least 100 particles. The Au–Pd NPs
were purified using the centrifuge method described in our
recent publication19 to obtain high-quality EXAFS data. The
Au L3-edge EXAFS measurements were conducted at the
HXMA beamline at the Canadian Light Source (CLS) oper-
ated at 2.9 GeV. Purified Au–Pd NPs were cast onto a My-
lar tape to form a uniform film, and the samples were
measured at room temperature using a transmission mode.
The EXAFS data were normalized and fitted with theoretical
phase-shift and amplitude generated from FEFF 8 program20
using a WinXAS program, following the procedure de-
scribed in our recent papers.19,21 The best-fit FT-EXAFS
was provided in the Supplementary data.
trated in Scheme 2 by inspecting the interaction between
DDAB and Au NP. The electronic effect in Finke’s model
can be described as the Coulombic repulsive interaction be-
tween NPs due to the binding of Br – onto Au NP surface,
resulting in negatively charged NPs. That is why the elec-
tronic effect for these ionic ligands is also called the charge
effect. The steric effect, on the other hand, originates from
the relatively bulky ammonium groups that stay near the
negatively charged NPs as counter cations. For DDAB used
In the present work, the two 12-carbon alkyl chains steri-
cally hinder direct contact between Au NPs.
When non-ionic ligands interact with Au NPs, the charge-
induced Coulombic effect used for ionic ligands does not di-
rectly apply. However, the steric effect alone is not sufficient
to account for the stabilities of NPs. For instance, Au NPs
stabilized with dodecylamine and dodecylthiol ligands should
experience very similar steric effects, but these two types of
NPs have significantly different stabilities (amine–Au <
thiol–Au). A second important stabilizing effect is apparently
related to the metal–ligand electronic bonding (e.g., Au–SR
and Au–amine), which can be considered as another type of
electronic effect. Such an electronic effect for non-ionic li-
gands can also play an important role in the stabilization of
NPs. In a later section, it will be shown that the electronic
and steric effects for non-ionic ligands, which are considered
as a simple extension of the models for ionic ligands, can be
used to understand the ligand-dependent NP size when Au
samples were capped with primary, secondary, tertiary
amines, and with amine/DDAB mixtures.
Au NPs in aqueous vs. toluene phase
Recently, it has been reported that Gemini surfactants
form a unique category of capping ligands for the prepara-
tion of metal nanoparticles in the aqueous phase.22,23 In ad-
dition, the molecular structures of the Gemini surfactants
can be altered to a much greater degree than those of the
monomers, which may offer new opportunities to control
the size and stability of NPs. Herein, two novel Gemini sur-
factants (Gem B and Gem C) together with a standard Gem-
ini surfactant (Gem A) were used for the synthesis of Au
NPs in aqueous solutions. The molecular structures of these
Gemini surfactants are displayed in Scheme 1. In addition,
DDAB was also used as a reference, since it has been found
that DDAB can form a bilayer capping structure similar to
those of the Gemini capping ligands.13 Figure 1 shows the
Results and discussion
The mechanism of nanoparticle stabilization by ligands
Finke and co-workers have recently reviewed the stabili-
zation modes of catalytic NPs with a particular emphasis on
metal NPs capped with anionic ligands.2,4 Electronic and
steric effects are proposed as the main factors influencing li-
gand stabilization. These two effects are schematically illus-
Published by NRC Research Press

1644
Can. J. Chem. Vol. 87, 2009
Fig. 1. TEM images of aqueous Au NPs capped with (a) DDAB,
(b) Gem A, (c) Gem B, and (d) Gem C. The average size and stan-
dard deviation are also presented in the figure.
Fig. 2. TEM images of toluene-phase Au NPs capped with (a)
RNH2, (b) R2NH, (c) R3N, and (d) DDAB.
distance from Au (e.g., amines with small-sized subsituents),
the electronic effects are identical. On the other hand, ionic
DDAB ligand should have different electronic effects from
the neutral amine ligands.
Figure 2 shows the TEM results of Au NPs prepared us-
ing DDAB and three amines. The sizes of RNH2-capped
(2.8 nm) and R2NH-capped (2.5 nm) NPs are very close to
each other. Based on the model illustrated in Scheme 2, this
observation can be easily understood. For RNH2- and R2NH-
capped NPs, the electronic effect is dominant; thus, the two
NP samples have very similar sizes due to their identical
metal–ligand bonding. On the other hand, the steric effect
should play a less significant role when going from a pri-
mary to a secondary amine.
Interestingly, the sizes of R3N-capped NPs are consider-
ably different from those of primary- and secondary-amine-
capped samples. Indeed, as seen in Fig. 2c, there exist two
groups of particles for R3N-capped NPs, one group being
3.7 nm in diameter and the other is 6.5 nm. Similar phenom-
ena were observed from other spots on the same TEM grid.
The steric effect for the bulky R3N ligand should be more
significant than the other two amines, favouring the forma-
tion of smaller NPs. However, it has been shown that due
to steric hindrance, the electronic bonding between Au and
amine is significantly weakened for bulky tertiary amines.15
As a result, the electronic effect for R3N ligands should be
weaker than that for RNH2 and R2NH ligands. Indeed, the
overall low monodispersity of R3N-capped NPs also sup-
ports the suggestion of a weaker electronic effect for the ter-
tiary ligands.
The DDAB-capped sample has very different particle size
(9.3 nm) from those capped with amines, which can be easily
attributed to the pronounced difference of its stabilization
mechanism from that of the neutral amine ligands. Indeed, it
has been reported that amines preferentially bind to low-co-
ordination Au sites (e.g., adatoms and corner atoms), whereas
those bound to terrace-type surface atoms are not stable.15,26
In contrast, tetraalkylammonium bromide molecules were
found to be mainly adsorbed to the terrace atoms on Au sur-
face, although the Br– alone, being separated from the ammo-
TEM images of four Au NP samples prepared with the four
ligands in water. It can be seen that all four capping ligands
lead to Au NPs of similar sizes (average diameter ranging
from 3.0 nm to 3.8 nm), although their molecular structures,
including both the binding groups and the steric non-binding
moiety, vary. It has been proposed that the double-chained
structure of surfactants facilitates the formation of
vesicles.13,23–25 The inner polar heads of the vesicles bind to
the metal surface and the outer ones are exposed to the
aqueous environment. The hydrophobic alkyl chains exist
inside bilayers with quite dense packing through chain–
chain interaction. Formation of such densely packed bilayers
should significantly enhance the role of the steric effect in
the stabilization of NPs and thus make it dominant in con-
trolling the NP size. The dominant role of the steric effect
can be more clearly revealed by the identical size of Au
NPs capped with Gem B and Gem C (3.8 nm vs. 3.6 nm),
whose binding head-groups are different (bromide vs. sul-
fate, respectively) but have similar steric moieties.
Next, we turn to toluene-phase Au NPs prepared with
DDAB and three fatty amines (primary, secondary, and terti-
ary) under identical synthetic conditions. The use of amines
with systematically varied structures in the synthesis of NPs
is particularly interesting in light of the recent remarkable
findings in amine-based single-molecule circuits. Venkatara-
man and co-workers reported that amines bind selectively to
undercoordinated Au atoms with sufficient angular flexibility
for easy junction formation but well-defined electronic cou-
pling of the N lone pair to the Au.16 It was also found that
the variability of observed conductance for diamine–Au
junctions is much less than that for dithiol–Au junctions.16
Therefore, amines can be considered as promising molecules
for the construction of single-molecule junctions and cir-
cuits. According to recent results from theoretical studies,
methyl-substituted primary, secondary, and tertiary amines
almost have the same adsorption energy when binding to
undercoordinated Au structures such as an adatom.26 These
results suggest that when the three amines stay at the same
Published by NRC Research Press

Murphy et al.
1645
Scheme 3. An illustrative presentation of the surface-capping structure of Au NPs in aqueous (a) and toluene phase (b).
nium counterions, can strongly bind to low-coordination Au
sites.27 Therefore, the considerable NP size increase when
going from amines to DDAB can be accounted for by consid-
ering their different ligand-binding mechanisms. The use of
amines resulted in much smaller NPs which contain higher
percentage of low-coordination surface atoms (i.e., adatoms,
corner, and edge atoms) that can be bound by the ligands; the
use of DDAB produced bigger NPs where the terrace-type
surface atoms are more abundant.
Au NPs with mixed ligands
To further study the interplay between the electronic and
steric effects of the ligands as well as to explore the possi-
bility of manipulating the two effects to control NP size, we
next used a combination of amine and DDAB to synthesize
Au NPs. Nine NP samples (3 Â 3) were prepared using
mixed ligands of RNH2, R2NH, or R3N with DDAB, each
series having 20%, 50%, and 80% amines, respectively. Fig-
ure 3 shows the TEM images of these nine NP samples.
UV–vis results of the three series of Au NPs (i.e., NPs
capped with DDAB/RNH2, DDAB/R2NH, and DDAB/R3N)
with systematically varied concentration of amines (0%,
20%, 50%, 80%, and 100%) are also displayed in the figure.
A few interesting observations are noted. Firstly, the three
samples capped with mixed DDAB and RNH2, shown in
Figs. 3a–3c, all have similar sizes (2.4 nm to 2.8 nm in di-
ameter) to that of pure RNH2-capped NPs (2.8 nm), which
are considerably different from that of DDAB-capped sam-
ples (9.3 nm). The UV–vis spectra in Fig. 3j show that all
the amine-capped NPs exhibit diminishing surface plasmon
resonance band, indicating the small sizes for these NPs
and consistently supporting the TEM data. Secondly, the
DDAB/R2NH-capped NPs displayed in Figs. 3d–3f exhibit
nearly the same particle size (6.8 nm to 7.1 nm in diameter),
which are very different from pure R2NH-capped Au sample
(2.5 nm), consistent with the UV–vis results in Fig. 3k.
Thirdly, the size of DDAB/R3N-capped NPs increase gradu-
ally with the concentration of DDAB, from 5.7 nm (80%
R3N) to 6.7 nm (50% R3N) and then to 7.6 nm (20% R3N).
The UV–vis spectra of NPs in this series in Fig. 3l all ex-
hibit fairly intense surface plasmon resonance bands. A
close inspection of Fig. 3l reveals that the surface plasmon
resonance band of Au shows a noticeable blue shift when
DDAB is present, even if the concentration of DDAB is
low (20%).
One may also consider the possibility that the size differ-
ence among amine- and DDAB-capped NPs observed in
Fig. 2 might be caused by different reduction rates of gold
compound in toluene phase by aqueous NaBH4. Since
DDAB is a phase-transfer agent, the absence of DDAB in
the three amine-capped samples may cause a slower reduc-
tion rate. However, the fact that Au NPs capped with 100%
RNH2 and those capped with a mixture of 80% RNH2 and
20% DDAB (see next section) have very similar sizes
(2.8 nm vs. 2.4 nm) suggests that such a possibility can be
ruled out. Note that the presence of Cl– anions from AuCl3
starting materials can also bind to the NP surface. However,
the possibility that the chloride anions play an important
role in determining the NP size can be ruled out in this
work. This argument is supported by the following two
facts. First, it is well-known that bromide (from DDAB)
can bind to gold surface more strongly than chloride.
Therefore, in the presence of DDAB, the anion adsorption
effect associated with Br – should be more significant. Sec-
ond, even in the absence of bromide, the above argument
seems still true as the Au NP size was found to be largely
determined by the molecular structure of amines (Fig. 2). It
is also noted that different sizes for DDAB-capped Au NPs
have been reported in the literature.28 This could be attrib-
uted to differences between the synthetic procedures. In-
deed, the stability of weakly bound NPs is largely kinetic in
nature, thus causing the NP size to vary in different re-
ports.29 In the present work, it is still valid to make a com-
parison between the amine- and DDAB-capped NPs, as they
were prepared under identical conditions.
The use of cosurfactants adsorbing onto Au 2D surfaces
has been reported.30 Generally, the existence of a cosurfac-
tant can either reduce the adsorption of the primary surfac-
tant due to a competitive adsorbing mechanism or enhance
the adsorption caused by a collaborative adsorption of the
two surfactants. The results of DDAB/RNH2-capped NPs in
Figs. 3a–3c suggest a competitive adsorbing process be-
tween DDAB and RNH2 ligands and indicate that stabiliza-
tion of Au NPs by the latter is dominant. A plausible reason
is that RNH2 is much smaller in size than DDAB, making
the former more competitive in the coadsorption process.
This is consistent with the reported results in the literature
Overall, a comparison of the syntheses of Au NPs in
aqueous and toluene phases highlights the very different
roles that the steric and electronic effects of the ligands
play, which is schematically illustrated in Scheme 3 using
DDAB as an example. Due to the existence of densely
packed bilayer capping structure in aqueous solutions, the
steric effect is dominant, whereas in toluene, the electronic
effect is more important than the steric effect.
Published by NRC Research Press

1646
Can. J. Chem. Vol. 87, 2009
Fig. 3. TEM images of Au NPs (in toluene) prepared using mixed amines and DDAB: (a) 80% RNH2, (b) 50% RNH2, (c) 20% RNH2, (d)
80% R2NH, (e) 50% R2NH, (f) 20% R2NH, (g) 80% R3N, (h) 50% R3N, and (i) 20% R3N. (j)–(l) are UV–vis spectra of RNH2 series (j),
R2NH series (k), and R3N series (l).
on Au 2D surface where cosurfactants of smaller sizes even
at low concentration (e.g., 1%–2%) greatly reduced the
amount of adsorbed primary surfactants.30 The results of
DDAB/R2NH-capped NPs, however, seem to be very differ-
ent from the case of primary amine. The sizes of the three
NP samples (~7 nm) shown in Figs. 3d–3f are close to that
of pure DDAB-capped NPs (9.3 nm), but quite different
from that of pure R2NH-stabilized sample (2.5 nm). Interest-
Published by NRC Research Press

Murphy et al.
1647
Fig. 4. TEM images of (a) Au0.25Pd0.75 NPs, (b) Au0.50Pd0.50 NPs,
(9.3 nm) and remained almost unchanged when the concen-
tration of DDAB was increased from 20% to 80%.
and (c) Au0.75Pd0.25 NPs. Au L3-edge FT-EXAFS is presented in (d)
–1
˚
using k-space EXAFS data of 3–13 A . Idealized local structure
In summary, the particle-size data of Au NPs capped with
mixed amines and DDAB ligands indicate that the coadsorp-
tion mechanism of ligands determines the NP size. RNH2 is
more competitive than DDAB in the coadsorption process;
thus, the NPs capped with 20%–80% RNH2 are all very
small (<3 nm), similar to that of pure RNH2-capped NPs.
R2NH may be coadsorbed onto Au in a collaborative way
with DDAB, making the NP size close to that of pure
DDAB-capped ones. R3N and DDAB, whose competitive
abilities in the coadsorption process are comparable, are sim-
ilarly important in determining the NP size. Consequently,
the NP size increases with the concentration of DDAB.
Overall, these results suggest that by selecting ligands with
desired molecular structure (and thus the electronic and
steric effect), one can possibly control Au NP size with this
mixed-capping method. Although further experimentation is
needed to rigorously evaluate the effectiveness of this
mixed-capping method, the results presented in Fig. 3 do
show some promise of more flexible control of NP size than
using a single type of ligands.
models of the three NPs are also shown in (d).
Au–Pd NPs with mixed ligands
We have recently reported the use of this mixed-capping
method to prepare Ag–Pd and Au–Pd NPs.19,21 Herein, we
present results on bimetallic Au–Pd NPs capped with mixed
DDAB (20%) and RNH2 (80%) ligands. These results can be
considered as an extension of the work on Au monometallic
NPs presented in the above sections. Moreover, the effect of
weakly binding ligands on the structure of bimetallic NPs is
even more interesting, as not only the particle size but also
the bimetal mixing pattern can possibly be influenced by
selecting the ligands. We have recently reported EXAFS
results on the local structure of Au–Pd NPs capped with a
mixture of 20% tetraoctylphosphonium bromide (TOPB)
and 80% n-dodecylamine, which show alloying-structure-
dependent physical and chemical properties.19 In the present
work, we intend to compare the local structure of Au–Pd
NPs capped with 20% DDAB and 80% n-dodecylamine
(RNH2) with the previous results with an attempt to study
the effect of different type of bromide ligands (DDAB vs.
TOPB) on the bimetal mixing pattern of Au–Pd NPs.
Figures 4a–4c depict the TEM images of Au–Pd NPs with
varied concentrations of Au. The average sizes for
Au0.75Pd0.25, Au0.50Pd0.50, and Au0.25Pd0.75 are 4.3 0.7 nm,
3.3 0.5 nm, and 3.7 1.0 nm, respectively. The fact that
these NPs are all bigger than their pure Au counterpart
(2.4 nm) implies that the metal–ligand binding differed after
Pd was introduced. To investigate the ligand effect on the
mixing patterns of Au–Pd NPs, Au L3-edge EXAFS meas-
urements were performed. The FT-EXAFS and fitting re-
sults were presented in Fig. 4d and Table 1, respectively. A
comparison of the experimental EXAFS data and first-shell
best fit is provided in the Supplementary data. The informa-
tion about Au–Pd mixing pattern can be obtained by analyz-
ing the coordination numbers (NAu–Au, NAu–Pd, and Ntotal) in
the table. It has been established19,31–33 that if NAu–Au/NAu–Pd
is equal to Au/Pd molar ratio, a homogeneous alloy was re-
sulted. If NAu–Au/NAu–Pd is considerably greater than Au/Pd
ingly, the size of the three DDAB/R2NH-capped NPs, which
are all slightly smaller than that of DDAB-capped ones,
does not change as the concentration of DDAB increases. A
possible explanation will be discussed in conjunction with
the discussion of DDAB/R3N-capped NPs. The results of
DDAB/R3N-capped NPs clearly indicate that both DDAB
and R3N ligands play an important role in the NP synthesis.
As a result, when more DDAB ligands were used, the NP
size was closer to that of pure DDAB-capped NPs. As was
discussed in a previous section, R3N has the weakest elec-
tronic effect among the three amine ligands. Therefore, it is
understandable that the adsorption of DDAB is significant
even at low concentration (20%). Supporting evidence is
provided by the UV–vis results presented in Fig. 3l. How-
ever, the secondary amine (R2NH), whose electronic effect
is stronger than the tertiary amine, seems to be less compet-
itive than R3N in the coadsorption process with DDAB.
Considering their very similar molecular structure (see
Scheme 1), we propose that R2NH and DDAB were collabo-
ratively adsorbed onto the NP surface.30 Since the NP surfa-
ces mainly consist of terrace atoms, it can be easily
understood that DDAB ligands are the ones that mainly de-
termine the NP size, as amines cannot form stable bonds
with the terrace Au atoms.26 This notion can also explain
why the particle size of DDAB/R2NH-capped samples
(~7 nm) is close to that of pure DDAB-capped NPs
Published by NRC Research Press

1648
Can. J. Chem. Vol. 87, 2009
Table 1. Au L3-edge EXAFS fitting results for three Au–Pd NP samples.
b
˚
Sample (% Au)
25
M–M bond Na
Ntotal
11.8
R (A)
Dd2
DE (eV)
0.88
2.28
–1.14
0.63
–3.29
3.28
Au–Au
Au–Pd
Au–Au
Au–Pd
Au–Au
Au–Pd
7.4
4.4
7.8
4.0
10.4
2.0
2.830
2.804
2.827
2.797
2.810
2.793
0.0088
0.0082
0.0090
0.0093
0.0088
0.0079
50
75
11.8
12.4
˚
Note: The uncertainty is ~20% for N and ~0.02 A for R.
aNearest-neighbor coordination number.
bFirst-shell metal–metal bond distance.
ligands (cosurfactants) were used. The experimental findings
presented in this work suggest that manipulation of the elec-
tronic and steric effects of weakly interacting ligands and
the use of mixed capping ligands can offer useful routes to
control NP size and bimetal mixing patterns.
molar ratio, then Au-core Pd-shell NPs were formed. More-
over, Ntotal is useful to obtain the information about the loca-
tion of Au atoms. If Ntotal is 12, which is the first-shell
coordination number for a bulk fcc metal, Au atoms cannot
be found on the NP surface (i.e., all the neighboring sites for
Au are occupied). If Ntotal is less than 12, some Au atoms
should be located on the surface, as the surface atoms of a
metal with fcc crystalline structure have a coordination num-
ber less than 12. The results of NAu–Au and NAu–Pd in Table 1
indicate that all the three Au–Pd NPs have core-shell struc-
tures. In addition, all the three Ntotal values are nearly 12, in-
dicating that the particle surfaces are essentially all covered
by Pd atoms. Based on the above analysis, the structural
models for the three Au–Pd NPs are schematically illus-
trated in Fig. 4. The core-shell structural models are consis-
tent with the TEM results, which show that Au–Pd NPs are
all bigger than pure Au NPs, since the surface metal–ligand
bonding in the Au-core Pd-shell NPs is different from that
of pure Au NPs. It is interesting to note that the present
structural results of Au–Pd NPs capped with DDAB/RNH2
(80% RNH2) differ from those capped with TOPB/RNH2
(80% RNH2) we recently reported.19 For the TOPB/RNH2-
capped NPs, core-shell structures were found only for the
Au0.50Pd0.50 and Au0.75Pd0.25 NPs, whereas the Au0.25Pd0.75
NPs showed a cluster-on-cluster structure. Therefore, a com-
parison of the bimetal mixing patterns of these two series of
samples indicates that the Au–Pd bimetallic mixing pattern
can be modified by varying the type of ligands used in this
mixed-capping method. We are in the process of systemati-
cally investigating the correlation between the Au–Pd mix-
ing pattern and the structure/concentration of the two types
of ligands (e.g., TOPB, DDAB, amines, and so forth) used
in the mixed-capping synthesis.
Supplementary data
Supplementary data for this article are available on the
journal Web site (canjchem.nrc.ca) or may be purchased
from the Depository of Unpublished Data, Document Deliv-
ery, CISTI, National Research Council Canada, Ottawa, ON
K1A 0R6, Canada. DUD 5287. For more information on ob-
taining material, refer to cisti-icist.nrc-cnrc.gc.ca/cms/
unpub_e.shtml.
Acknowledgements
We thank the financial support for this work by Dalhousie
University, the Natural Sciences and Engineering Research
Council of Canada (NSERC), the Atlantic Innovation Fund,
and Canada Foundation for Innovation (CFI). The CLS is
funded by NSERC, the Canadian Institutes of Health Re-
search (CIHR), the National Research Council Canada
(NRC), and the University of Saskatchewan. The EXAFS
technical support from CLS staff scientist, Dr. Ning Chen,
is also acknowledged.
References
(1) Templeton, A. C.; Wuelfing, W. P.; Murray, R. W. Acc.
Chem. Res. 2000, 33 (1), 27–36. doi:10.1021/ar9602664.
PMID:10639073.
(2) Finke, R. G.; Ozkar, S. Coord. Chem. Rev. 2004, 248 (1-2),
135–146. doi:10.1016/j.cct.2003.08.003.
(3) Daniel, M. C.; Astruc, D. Chem. Rev. 2004, 104 (1), 293–
346. doi:10.1021/cr030698+. PMID:14719978.
(4) Ott, L. S.; Finke, R. G. Coord. Chem. Rev. 2007, 251 (9-10),
1075–1100. doi:10.1016/j.ccr.2006.08.016.
(5) Sperling, R. A.; Rivera Gil, P.; Zhang, F.; Zanella, M.;
Parak, W. J. Chem. Soc. Rev. 2008, 37 (9), 1896–1908.
doi:10.1039/b712170a. PMID:18762838.
(6) Brust, M.; Fink, J.; Bethell, D.; Schiffrin, D. J.; Kiely, C. J.
Chem. Soc. Chem. Commun. 1995, 10 (16), 1655. doi:10.
1039/c39950001655.
Conclusion
To conclude, we have reported results of Au NP synthesis
using a variety of weakly interacting ligands. Aqueous Au
NPs capped with DDAB and three Gemini surfactants were
found to have very similar sizes (average diameter of 3–
4 nm), whereas in toluene, the use of DDAB, amines, and
their mixtures as capping ligands led to Au NPs with aver-
age sizes ranging from 2.5 nm to 9.3 nm. The ligand effect
on the bimetal mixing patterns of Au–Pd NPs was also
studied with EXAFS. These results can be accounted for by
considering the electronic and steric effects of capping li-
gands and the ligand coadsorption mechanism when mixed
(7) Hostetler, M. J.; Templeton, A. C.; Murray, R. W. Langmuir
1999, 15 (11), 3782–3789. doi:10.1021/la981598f.
(8) Song, Y.; Murray, R. W. J. Am. Chem. Soc. 2002, 124 (24),
7096–7102. doi:10.1021/ja0174985. PMID:12059234.
Published by NRC Research Press

Murphy et al.
1649
(9) Song, Y.; Huang, T.; Murray, R. W. J. Am. Chem. Soc.
2003, 125 (38), 11694–11701. doi:10.1021/ja0355731.
PMID:13129374.
(20) Ankudinov, A. L.; Ravel, B.; Rehr, J. J.; Conradson, S. D.
Phys. Rev. B 1998, 58 (12), 7565–7576. doi:10.1103/
PhysRevB.58.7565.
(10) Roucoux, A.; Schulz, J.; Patin, H. Chem. Rev. 2002, 102
(10), 3757–3778. doi:10.1021/cr010350j. PMID:12371901.
(11) Murphy, C. J.; Sau, T. K.; Gole, A. M.; Orendorff, C. J.;
Gao, J. X.; Gou, L.; Hunyadi, S. E.; Li, T. J. Phys. Chem. B
2005, 109 (29), 13857–13870. doi:10.1021/jp0516846.
PMID:16852739.
(12) Fink, J.; Kiely, C. J.; Bethell, D.; Schiffrin, D. J. Chem. Ma-
ter. 1998, 10 (3), 922–926. doi:10.1021/cm970702w.
(13) Zhang, L. X.; Sun, X. P.; Song, Y. H.; Jiang, X.; Dong, S. J.;
Wang, E. A. Langmuir 2006, 22 (6), 2838–2843. doi:10.
1021/la052822l. PMID:16519492.
(21) Cook, S. C.; Padmos, J. D.; Zhang, P. J. Chem. Phys. 2008,
128 (15),o. 154705. doi:10.1063/1.2901034. PMID:18433256.
(22) Liu, Q.; Guo, M. L.; Nie, Z.; Yuan, J. B.; Tan, J.; Yao, S. Z.
Langmuir 2008, 24 (5), 1595–1599. doi:10.1021/la702978z.
PMID:18237211.
(23) Zhong, L.; Jiao, T.; Liu, M. H. Langmuir 2008, 24 (20),
11677–11683. doi:10.1021/la802338f. PMID:18823092.
(24) Zana, R.; Talmon, Y. Nature 1993, 362 (6417), 228–230.
doi:10.1038/362228a0.
(25) Danino, D.; Talmon, Y.; Zana, R. Langmuir 1995, 11 (5),
1448–1456. doi:10.1021/la00005a008.
(14) Grubbs, R. B. Pol. Rev. 2007, 47 (2), 197–215. doi:10.1080/
15583720701271245.
(15) Venkataraman, L.; Klare, J. E.; Nuckolls, C.; Hybertsen, M.
S.; Steigerwald, M. L. Nature 2006, 442 (7105), 904–907.
doi:10.1038/nature05037. PMID:16929295.
(16) Venkataraman, L.; Klare, J. E.; Tam, I. W.; Nuckolls, C.;
Hybertsen, M. S.; Steigerwald, M. L. Nano Lett. 2006, 6
(3), 458–462. doi:10.1021/nl052373+. PMID:16522042.
(17) Quek, S. Y.; Venkataraman, L.; Choi, H. J.; Louie, S. G.;
Hybertsen, M. S.; Neaton, J. B. Nano Lett. 2007, 7 (11),
3477–3482. doi:10.1021/nl072058i. PMID:17900162.
(18) (a) Jahan, N.; Paul, N.; Petropolis, C. J.; Marangoni, D. G.;
Grindley, T. B. J. Org. Chem., 2009, 74 (20), 7762–7773.
doi:10.1021/jo9018107.; (b) Tran, T.; Jahan, N.; Marangoni,
D. G.; Grindley, T. B. Manuscript in preparation.
(26) Hoft, R. C.; Ford, M. J.; McDonagh, A. M.; Cortie, M. B. J.
Phys. Chem. C 2007, 111 (37), 13886–13891. doi:10.1021/
jp072494t.
´
(27) Kiraly, Z.; Findenegg, G. H.; Mastalir, A. Langmuir 2006,
22 (7), 3207–3213. doi:10.1021/la053184+. PMID:16548579.
(28) Lin, X. M.; Sorensen, C. M.; Klabunde, K. J. Chem. Mater.
1999, 11 (2), 198–202. doi:10.1021/cm980665o.
(29) Leff, D. V.; Brandt, L.; Heath, J. R. Langmuir 1996, 12 (20),
4723–4730. doi:10.1021/la960445u.
(30) Knag, M.; Sjoblom, J.; Gulbrandsen, E. J. Disper. Sci. Tech.
(Paris) 2005, 26 (2), 207–215. doi:10.1081/DIS-200045595.
(31) Frenkel, A. I.; Hills, C. W.; Nuzzo, R. G. J. Phys. Chem. B
2001, 105 (51), 12689–12703. doi:10.1021/jp012769j.
(32) Davis, R. J.; Boudart, M. J. Phys. Chem. 1994, 98 (21),
5471–5477. doi:10.1021/j100072a012.
(19) Liu, F.; Wechsler, D.; Zhang, P. Chem. Phys. Lett. 2008, 461
(4-6), 254–259. doi:10.1016/j.cplett.2008.07.029.
(33) Harada, M.; Asakura, K.; Toshima, N. J. Phys. Chem. 1993,
97 (19), 5103–5114. doi:10.1021/j100121a042.
Published by NRC Research Press