A scalable synthesis of highly stable and water dispersible Ag 44(SR)30 nanoclusters

Article abstract of DOI:10.1039/c3ta11785e

We report the synthesis of atomically monodisperse thiol-protected silver nanoclusters [Ag₄₄(SR)₃₀] ^m, (SR = 5-mercapto-2-nitrobenzoic acid) in which the product nanocluster is highly stable in contrast to previous preparation methods. The method is one-pot, scalable, and produces nanoclusters that are stable in aqueous solution for at least 9 months at room temperature under ambient conditions, with very little degradation to their unique UV-Vis optical absorption spectrum. The composition, size, and monodispersity were determined by electrospray ionization mass spectrometry and analytical ultracentrifugation. The produced nanoclusters are likely to be in a superatom charge-state of m = 4-, due to the fact that their optical absorption spectrum shares most of the unique features of the intense and broadly absorbing nanoparticles identified as [Ag₄₄(SR) ₃₀]^4- by Harkness et al. (Nanoscale, 2012, 4, 4269). A protocol to transfer the nanoclusters to organic solvents is also described. Using the disperse nanoclusters in organic media, we fabricated solid-state films of [Ag₄₄(SR)₃₀]^m that retained all the distinct features of the optical absorption spectrum of the nanoclusters in solution. The films were studied by X-ray diffraction and photoelectron spectroscopy in order to investigate their crystallinity, atomic composition and valence band structure. The stability, scalability, and the film fabrication method demonstrated in this work pave the way towards the crystallization of [Ag₄₄(SR)₃₀]^m and its full structural determination by single crystal X-ray diffraction. Moreover, due to their unique and attractive optical properties with multiple optical transitions, we anticipate these clusters to find practical applications in light-harvesting, such as photovoltaics and photocatalysis, which have been hindered so far by the instability of previous generations of the cluster.

Full text of DOI:10.1039/c3ta11785e

Journal of
Materials Chemistry A
View Article Online
View Journal | View Issue
PAPER
A scalable synthesis of highly stable and water
dispersible Ag (SR) nanoclusters†
4
4
30
Cite this: J. Mater. Chem. A, 2013, 1,
0148
1
a
b
a
c
Lina G. AbdulHalim,‡ Sumaira Ashraf,‡ Khabiboulakh Katsiev, Ahmad R. Kirmani,
d
e
f
c
Nuwan Kothalawala, Dalaver H. Anjum, Sikandar Abbas, Aram Amassian,
g
d
f
a
Francesco Stellacci, Amala Dass, Irshad Hussain* and Osman M. Bakr*
m
We report the synthesis of atomically monodisperse thiol-protected silver nanoclusters [Ag44(SR)30 ] , (SR ¼ 5-
mercapto-2-nitrobenzoic acid) in which the product nanocluster is highly stable in contrast to previous
preparation methods. The method is one-pot, scalable, and produces nanoclusters that are stable in
aqueous solution for at least 9 months at room temperature under ambient conditions, with very little
degradation to their unique UV-Vis optical absorption spectrum. The composition, size, and monodispersity
were determined by electrospray ionization mass spectrometry and analytical ultracentrifugation. The
produced nanoclusters are likely to be in a superatom charge-state of m ¼ 4ꢀ, due to the fact that their
optical absorption spectrum shares most of the unique features of the intense and broadly absorbing
4ꢀ
nanoparticles identified as [Ag44(SR)30
]
by Harkness et al. (Nanoscale, 2012, 4, 4269). A protocol to
transfer the nanoclusters to organic solvents is also described. Using the disperse nanoclusters in organic
m
media, we fabricated solid-state films of [Ag44(SR)30
]
that retained all the distinct features of the optical
absorption spectrum of the nanoclusters in solution. The films were studied by X-ray diffraction and
photoelectron spectroscopy in order to investigate their crystallinity, atomic composition and valence band
structure. The stability, scalability, and the film fabrication method demonstrated in this work pave the way
Received 6th May 2013
Accepted 25th June 2013
m
towards the crystallization of [Ag44(SR)30
]
and its full structural determination by single crystal X-ray
diffraction. Moreover, due to their unique and attractive optical properties with multiple optical transitions,
we anticipate these clusters to find practical applications in light-harvesting, such as photovoltaics and
photocatalysis, which have been hindered so far by the instability of previous generations of the cluster.
DOI: 10.1039/c3ta11785e
www.rsc.org/MaterialsA
emerging family of model compounds within the technologi-
cally important class of thiolate protected metal nanoparticles.
Introduction
Atomically monodisperse thiolate (–SR) protected noble metal Their well-dened structures and discrete size distributions
1,2
nanoclusters (NCs, also known as nanomolecules ) are an have enabled researchers to envision and engage in profound
3
studies that elucidate their chemical and physical properties:
4
5,6
size-dependent catalysis, quantum-to-plasmon transition,
a
Division of Physical Sciences and Engineering, Solar and Photovoltaics Engineering
Center, King Abdullah University of Science and Technology (KAUST), Thuwal
7
8
chirality, magnetism, and the nature of the thiol–metal
interface. The central factor permitting such a leap in under-
9
2
3955-6900, Saudi Arabia. E-mail: osman.bakr@kaust.edu.sa
b
Department of Metallurgy & Materials Engineering, Pakistan Institute of Engineering standing is the crystallization of NCs into true macroscopically
&
Applied Sciences (PIEAS), Nilore, Islamabad, Pakistan
sized crystals and their full structural (metal core and ligand
shell) determination by single crystal X-ray diffraction. Typi-
cally, stable NCs of high purity are a prerequisite for realizing
crystals with the necessary quality for full structural determi-
c
Materials Science and Engineering Program, Division of Physical Sciences and
Engineering, King Abdullah University of Science and Technology (KAUST), Thuwal
2
d
3955-6900, Saudi Arabia
Department of Chemistry and Biochemistry, University of Mississippi, University,
Mississippi 38677, USA
e
nation. To date, the structures of an impressive number of gold
Advanced Nanofabrication Imaging and Characterization, King Abdullah University of thiolate NCs have been reported. The synthetic methods to
1,3
Science and Technology, Thuwal 23955-6900, Saudi Arabia
produce atomically monodisperse, yet stable, Au NCs reported
f
Department of Chemistry, SBA School of Science & Engineering (SSE), Lahore
University of Management Sciences (LUMS), DHA, Lahore Cantt – 54792, Pakistan.
E-mail: ihussain@lums.edu.pk
3
by several groups remain the key to such progress.
In contrast, the full structural characterization of only a few
10,11
silver thiolate NCs has so far been achieved.
Although
g
Institute of Materials, Ecole Polytechnique F ´e d ´e rale de Lausanne, Switzerland
multiple notable reports on the syntheses of silver thiolate NCs
†
1
‡
Electronic supplementary information (ESI) available. See DOI:
0.1039/c3ta11785e
12–22
have surfaced,
most of the products are either polydisperse
These authors contributed equally to this work.
or unstable, hindering efforts to crystallize them. Of all the
1
0148 | J. Mater. Chem. A, 2013, 1, 10148–10154
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
4
ꢀ
25
silver NCs reported so far, [Ag44(SR)30
]
(also known as intense were calculated using Ultrascan III (version 1.0, revision 1349),
14,23
23
broadly absorbing nanoparticles, IBANs)
interest not only as a model compound, but also for its rare and calculating s and D, the molecular weight was determined based
attractive optical properties: multiple strong absorption peaks on the work of Carney et al.
that cover a broad range of wavelengths (350–950 nm), and an Transmission electron microscopy (TEM) images were
unusually long-lived charge-separated excited state. These obtained using a TitanG2 80-300 instrument, equipped with an
is of special utilizing a procedure similar to the previous work. Aer
26
24
4
ꢀ
properties make [Ag44(SR)30
]
a potentially ideal material for image-corrector from CEOS. An energy-lter from Gatan Inc.
light-harvesting in photovoltaics and photocatalysis, should its was used at an acceleration voltage of 300 kV and an energy-slit
stability be considerably enhanced. Unfortunately, all previ- of width 10 eV to reduce the amount of inelastically scattered
4
ꢀ
ously published reports on [Ag (SR) ] have reported NCs electrons for improving the image contrast in high resolution
4
4
30
that are easily prone to oxidation at room temperature in a (HR) TEM micrographs of these NCs.
14,23,24
matter of hours.
Large angle powder X-ray diffraction (XRD) measurements
Here we report the synthesis of highly stable and atomically were performed on a Bruker D8 Advance with a Cu source. A
m
monodisperse [Ag44(SR)30
]
NCs (with a superatom charge- dispersion of Ag44(SR)30 NCs in methanol was drop-cast on a
state, m, most probably equal to 4ꢀ). Moreover, the synthesis is silicon single crystal substrate to avoid any signal contribution
simple and scalable. By using 5-mercapto-2-nitrobenzoic acid as from the background.
a ligand, we demonstrate that introducing nitro and carboxylic
X-ray photoelectron spectroscopy (XPS) measurements were
acid functionality in the aryl ligands not only renders the NCs done on a UHV Omicron surface analysis system, equipped with
water-soluble, but also stable in solution for a period of at least a SPHERA U7 hemispherical energy analyser with a 7-channel
9
months. We show that the NCs can be transferred into organic MCD detector, high intensity monochromated Al Ka X-rays and
solvents, and use them to fabricate solid lms that retain the He I radiation sources, and an XM 1000 monochromator.
m
unique absorption spectrum of [Ag44(SR)30] . The stability of
The ultraviolet photoemission spectroscopy (UPS) study was
the NCs facilitated their characterization in solution and in carried out at the Canadian Light Source. The photon energy
solid-state. Our work enables the practical consideration of used for this study was 36 eV with a total energy resolution of
4
ꢀ
[Ag44(SR)30
]
(hereaer referred to simply as Ag44(SR)30) in 30 meV. All spectra were collected by a Scienta SES100 spec-
photovoltaics, photocatalysis, and is a milestone towards its trometer at normal emission and room temperature.
crystallization and full structural determination.
The UV-Vis absorption spectra of silver NCs in solution were
recorded using a JAZZ spectrophotometer (Ocean Optics).
Experimental
3
6 4 4
Synthesis and purication of Ag44(SC H O N)30 NCs
1
Chemicals
0
In a 25 mL Erlenmeyer ask, DTNBA (9.91 mg, 25 mmol) was
stirred in NaOH aqueous solution (20 mL, 1 M). The orangish
solution turned to dark yellow indicating the cleavage of the
disulde bond yielding 5-mercapto-2-nitrobenzoic acid
All chemicals including 5,5 -dithiobis (2-nitrobenzoic acid)
(
DTNBA, 99%), silver nitrate (AgNO , 99%), sodium hydroxide
NaOH), sodium borohydride (NaBH , 99.99% metals basis), and
tetramethylammonium hydroxide (TMAH, 23%) were purchased
from Sigma Aldrich and used without further purication.
3
(
4
(
MNBA). An aqueous solution of AgNO (8.5 mg, 50 mmol, 5 mL)
3
was then added to the MNBA solution. The colour turned
greenish yellow indicating the formation of an Ag–S complex. A
2
Characterization techniques
fresh aqueous solution of NaBH
Nuclear magnetic resonance (NMR) spectra were recorded at reduce the complex. The solution turned dark brown immedi-
98 K on a Bruker 600 AVANAC III spectrometer equipped with a ately and gradually changed to dark red under vigorous stirring
Bruker BBO multinuclear probe. for 4 h, indicating the formation of Ag44(SR)30 NCs. The clusters
4
(1 mg, 2 mL) was then used to
2
Mass spectra were collected using a Waters Synaptꢀ G1 Q- were puried by repeated precipitation with 50% methanol
Tof High Denition Mass Spectrometer. The NC sample was followed by repeated centrifugation at 9000 rpm for 10 min and
dispersed in methanol (HPLC grade) and the solution was decantation of the supernatant until it became colourless.
electrosprayed in negative mode by using a stainless steel
syringe. The instrument parameters were maintained as
4
Film fabrication
follows: mass range 400 to 4000 Da; capillary voltage ꢀ2.1 kV;
detector voltage 1.7 kV; extraction cone voltage 2.1 V; sampling Films of clusters were drop-cast from methanol. In order to
ꢁ
cone voltage 10.0 V; source temperature 90 C; desolvation disperse the NCs in methanol, the puried NCs in water were
ꢁ
temperature 200 C; optics mode V. All the mass spectral data precipitated one more time with methanol, and then mixed
were processed using Masslynx 4.1 soware (Waters Corp.).
with 500 mL of TMAH. The solvent (water to methanol) exchange
Sedimentation velocity-analytical ultracentrifugation (SV- was established by concentrating the particle solution to 10% of
AUC) experiments were performed using an Optima XL-A its original volume in centrifuge lters (10 kDa M cut-off) and
w
analytical ultracentrifuge from Beckman Coulter with an reconstituting the rest of the volume with pure methanol. This
absorbance optical detection system and an An-60 Ti rotor. The cycle was repeated 3–4 times to obtain a stable solution of the
sedimentation and diffusion coefficient (s and D) distributions clusters in methanol with trace amounts of water.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 10148–10154 | 10149

View Article Online
Journal of Materials Chemistry A
Paper
work were colloidally and chemically stable for at least 9
months under ambient conditions as shown in Fig. 1a. The
choice of MNBA as a ligand was guided by our previous obser-
Results and discussion
1
Synthesis
4ꢀ
Recently, thiolated [Ag (SR) ]
NCs protected by 4-uoro- vation that the stability of the formed clusters was dependent
4
4
30
thiophenol, 3-uorothiophenol, 2-uorothiophenol, 4-mercapto- on the functional groups on the phenyl ring of the aryl ligands.1
4
benzoic acid, (2-mercapto-5-methylphenyl)methanol, and In particular, 4-mercaptobenzoic acid (4-MBA)-stabilized NCs
14
14
4
-methylbenzene-1,2-dithiol have been reported. These clusters were more stable than the 4-uorothiophenol (4-FTP) NCs.
exhibited unique absorption in the visible and near-IR regions of This suggested that the stability of these clusters might be due
14,23
the optical spectrum.
The exact molecular formula and the to the electron withdrawing groups (EWGs) of the ligand. In
monodispersity of these NCs were determined by ESI-MS and order to test this hypothesis, we investigated the use of 3-mer-
23
SV-AUC. Major shortcomings of the reported synthesis protocol captobenzoic acid (3-MBA) as a ligand, while the stability of the
of these NCs were their instability (i.e. growth into larger particles formed clusters was monitored by UV-Vis absorption spectros-
in a few hours at room temperature), easier oxidation, cumber- copy (see Fig. S3†). These studies indicated that the resulting
ꢁ
some purication, and being only stable if refrigerated at ꢀ4 C. clusters were stable for 3 to 4 days under ambient conditions at
Herein, we report the synthesis of Ag44(SR)30 NCs (with most basic pH.
probably a superatom charge of 4ꢀ, vide infra) that are stable for
We experimented with various derivatives of MBA and
at least 9 months at room temperature and under ambient found that MNBA (Scheme 1, step 2), which contained a nitro
conditions in aqueous media. This facilitates their character- group at the para position of the phenyl ring with respect to
ization and paves the way for their use in practical applications.
In this work, we used 5-mercapto-2-nitobenzoic acid (MNBA) as
a ligand for Ag44(SR)30 NCs. By using stabilizing ligands with
nitro and carboxylate electron withdrawing groups, we obtained
highly stable NCs with the same optical properties and molec-
ular assignment as previously reported Ag44(SR)30, yet stable for
several months at room temperature and under ambient
conditions. The structure of the ligand and the reaction scheme
are depicted in Scheme 1. Briey, the cleavage of the disulde
0
bond of 5,5 -dithiobis (2-nitrobenzoic acid (DTNBA)) was ach-
ieved by vigorous stirring in 1 M aqueous NaOH solution for
around 1 h to form MNBA. Subsequently, an equimolar solution
of Ag salt in water was added and the reaction mixture was
stirred at room temperature for another hour to ensure the
4
formation of an Ag–S complex. Finally, NaBH was added and
the solution was le overnight under vigorous stirring. The
resulting crude nanoclusters showed at least ve distinct
absorption peaks in the UV-visible-NIR region with absorption
maxima at 400, 480, 550, 650, and 850 nm (see the black curve in
Fig. S1†). These absorption peaks became even sharper and
more pronounced aer the purication step (see the red curve
in Fig. S1†). A typical reaction yielded approximately 4.3 mg
(50% conversion of Ag from AgNO
3
) of NCs. However, the
reaction can be scaled up by at least a factor of 40 (see Fig. S2†).
2
Stability
In contrast to the previous reports on Ag44(SR)30 that were stable
for only a few hours, the MNBA-stabilized NCs reported in this
Fig. 1 UV-Vis spectra of (a) Ag44(MNBA)30 NCs in 1 M NaOH aqueous solution
4 hours (black), 9 months (red) after synthesis; and (b) NCs transferred into
methanol (black) and in a solid form (red).
2
Scheme 1 Synthesis of Ag44(SR)30 nanoclusters.
1
0150 | J. Mater. Chem. A, 2013, 1, 10148–10154
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
thiol and adjacent to the carboxylic acid group, provided the
best stability, likely due to the combined effect of two EWGs.
Additionally, both groups are relatively hydrophilic, and
thereby enhanced the colloidal stability of the resulting NCs
in aqueous media. It is worth noting that when 4-nitrobenzoic
acid was used as a ligand, the resulting clusters were highly
unstable with aggregation quickly occurring aer synthesis,
suggesting that the electron withdrawing capabilities of the
nitro group is not enough to endow the NCs with the desired
stability.
We were also able to transfer the NCs to methanol using
TMAH (vide supra) as a positively charged counterion to the
14,23
cluster, which is overall negatively charged,
and used the
resulting solution to fabricate solid lms. The use of volatile
organic solvents aids in the fabrication of smooth lms using
solution based methods such as drop-casting or spin-coating.
The NCs in both methanol solution and solid-state retained the
Fig. 2 Negative ion ESI-MS of Ag44(MNBA)30 NCs in comparison with the
theoretical spectrum of Ag44(MNBA)30. The inset shows the expansion of the 5-
overall unique absorption spectrum of the original synthesized charge species.
clusters (see Fig. 1b).
To the best of our knowledge, this is one of the rst reports on
Ag44(SR)30 NCs (vide infra) that are stable, scalable and can be
easily puried. The stability of the cleaned clusters facilitated
their further characterization in liquid solution and in solid-state.
3
Size and composition
The size, composition, and chemical structure of the resulting
Ag NCs were investigated using a combination of different
analytical techniques including ESI, AUC, HRTEM, Powder
1
XRD, XPS, and H NMR. The results from these studies are
discussed below.
ESI-MS has become a potent technique used in measuring
23,27
the molecular weight of noble metal thiolate NCs.
The full
spectrum of the resulting Ag44(SR)30 NCs, measured in negative
mode, is shown in Fig. 2 together with the theoretical spectrum
for Ag44(SR)30. As apparent from Fig. 2, the four major peaks
corresponding to 4, 5, 6 and 7 charge states of the resulting NCs
are in a good agreement with their corresponding theoretical
values. The inset of Fig. 2 shows the expansion of the 5-charge
state centered at m/z 2165 Da, from which the molecular weight
of the resulting NCs was calculated to be ca. 10.8 kDa. A
common problem in mass spectrometric characterization of
metal nanoclusters is that they oen undergo fragmentation
during the process of ionization. We can exclude this possibility comprised of one species with s ¼ 3.57 ꢂ 10
Fig. 3 2D plot of sedimentation coefficients (s) and diffusivities (D) of the NCs.
ꢀ
13
s. The diffusion
ꢀ
6
2
ꢀ1
because the MS results are in close agreement with those of coefficient (D) corresponding to this species is 1.24 ꢂ 10 cm s
.
ꢀ
13
analytical ultracentrifugation (vide infra).
The remaining species at around s ¼ 3.3 ꢂ 10
s are likely the
SV-AUC is a complementary characterization technique to same species that did not sediment ideally due to particle–
TEM and mass spectrometry and is used to determine the particle interaction in solution. For the most abundant species,
homogeneity and approximate size of macromolecules and we found that the molecular weight is ca. 11.1 kDa. SV-AUC data
nanoparticles in solution. In previous reports we successfully on the clusters in solution are atomically monodisperse and
used SV-AUC to determine the approximate molecular weight, corroborate the mass spec assignments of Ag44(SR)30
density, and size distribution of Au and Ag NCs, by measuring HR-TEM was used to provide the insights into the core size
their sedimentation (s) and diffusion (D) coefficient distribu- and morphology of the Ag44(SR)30 NCs. In general, TEM cannot
tions in solution. be used to determine the exact structure of clusters in this size
The s and D distributions of the synthesized Ag (SR) NCs range and is also prone to imaging and sample preparation
.
3
4
4
30
in 1 M NaOH aq. are shown in Fig. 3. The distributions show artefacts because (i) the electron beam may alter the clusters;
that the clusters are homogeneous; at least 94% of the sample is (ii) the low-contrast of the smallest clusters results in a bias in
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 10148–10154 | 10151

View Article Online
Journal of Materials Chemistry A
Paper
Fig. 4 (a) TEM image of the synthesized NCs. (b) A magnified HR-TEM image of a
single cluster.
the size statistics and (iii) different orientations of the same
cluster are hard to distinguish from the heterogeneous
distribution of clusters. Nevertheless, from TEM images, one
can see that the particle shapes are not spherical (Fig. 4a) and Fig. 6 XPS survey of Ag44(MNBA)30 NCs.
that they have crystalline Ag cores (Fig. 4b), with a diameter
size in the range of ca. 1.5–2.5 nm.
1
63 eV; it is close to the reported value of metal sulphides
Powder XRD. For more accurate core size determination of
the cluster, we performed large-angle powder XRD (see Fig. 5).
The broad peak shown in Fig. 5 overlaps with the (111) and (200)
reections of bulk fcc-Ag. Thus we t two Gaussian peaks
around these positions and used Scherrer's equation to esti-
mate the crystal size. We found that the core size is 1.39 nm,
which is in good agreement with the core sizes inferred from
ESI-MS and SV.
29
(161.1–162.2 eV).
High-resolution XPS measurements were performed in order
to quantitatively analyse the elemental composition of the
Ag (SR) NCs. The integration of the area under Ag 3d and S 2p
peaks gave 43.3% Ag and 56.7% S atomic concentration, and in
reasonably good match with the values calculated for
Ag44(SR)30, where there are 40.5% Ag and 59.5% S.
4
4
30
28
1
1
H NMR. High-resolution H NMR was used to investigate
The XPS survey spectrum of Ag44(SR)30 NCs drop-cast on the
glass/ITO substrate is shown in Fig. 6. The spectrum reveals the
presence of Ag, S, C, N, O, and a residual amount of Na; no other
elements were detected. The Ag 3d5/2 peak is at 366.9 eV. This
value is 1.1 eV lower in binding energy (BE) compared to the
reported value for bulk Ag (368 eV). As for sulfur, the S 2p peak is
observed at 161.3 eV. This value does not correspond either to
free thiols or to disuldes, whose expected BEs are higher than
the ligand–shell structure of the nanoclusters. NMR spectros-
copy has previously shown great promise as a tool to under-
stand the chemical environments and the morphology at the
30–32
metal–ligand interface.
Fig. 7 shows the aromatic region of
1
the H NMR spectra of the ligand and of a dispersion of
Ag44(SR)30 NCs.
Fig.
5
Powder X-ray diffraction of the (111) and (200) reflections of
ꢀ1
1
Ag44(MNBA)30 NCs (black) plotted versus s ¼ 2 sin q/l in 1 nm . The bars (at the
bottom) correspond to the reflections of bulk fcc-Ag. The greenish curves corre-
spond to the Gaussian fitting of the spectrum to calculate the FWHM of the two
peaks while the red curve is the cumulative peak fit.
Fig. 7 Aromatic region of H NMR spectra of (a) free starting material ligands;
the inset shows the chemical structure of DTNBA and (b) Ag44(MNBA)30 NCs
dispersed in 1 M NaOH in D
2
O; integration of the cluster's protons are displayed
below their respective peaks.
1
0152 | J. Mater. Chem. A, 2013, 1, 10148–10154
This journal is ª The Royal Society of Chemistry 2013

View Article Online
Paper
Journal of Materials Chemistry A
0
By inspection of the NMR spectra of the starting material (5,5 -
dithiobis (2-nitribenzoic acid)) and the formed Ag (MNBA)
30
4
4
NCs, it is clear that all the peaks of the ligands attached to the
NCs are shied upeld. Such a shi is usually related to the
different surrounding environment of the ligands. The NMR
spectrum of the resulting NCs showed the presence of two
different ligand environments with the ratio of 3 : 1 (see inte-
gration of proton peaks in Fig. 7b). One proton set has a singlet at
7.65 and doublets at 7.18 and 7.2, while the other one has a
singlet at 7.45 and two doublets at 6.3 and 6.6. The observed
upeld shis are indicative of the shielding effect of the clusters
33
1
on the ligand protons. The full H NMR spectrum of the NCs
along with the chemical shi assignments is shown in Fig. S4.†
4
Optical and electronic characteristics
The UV-Vis absorption spectrum of the NCs shows at least ve
distinct absorption peaks at ꢃ400, ꢃ480, ꢃ550, ꢃ650 and
ꢃ850 nm, similar to the reported spectrum of the [Ag44(4-
Fig. 9 Normal emission UPS spectra of Ag44(MNBA)30 NCs (red curve) in
comparison with the Ag single crystal (black curve). The inset shows a zoomed-in
around B.E. ¼ 0.
4
ꢀ
14,23
FTP)30
]
NCs,
(see Fig. 8). Multiple absorption peaks in the
visible-NIR region are indicative of the molecular-like electronic
structure. It is well known that the optical absorption transi-
tions of discrete noble metal NCs in this size range, which
the spectrum of the Ag single crystal and consists of the seven
features at 2.8, 3.9, 5.3, 6, 7.5, 10.3 and 11.7 eV. The spectrum of
the NCs (up to some extent) resembles the valence band struc-
ture of the bulk Ag that was measured on the Ag (111) single
crystal and denoted as “Ag single crystal”. The valence band
structure of the bulk Ag shows the Ag d-band with the d-band
centered around 5.1 eV and the sharp Fermi edge (inset in
34
behave like superatom complexes, are sensitive to the overall
35
charge of the cluster. Indeed, one can expect the visible and
4
ꢀ
NIR optical transitions of an [Ag44(SR)30
]
cluster to be even
more sensitive to perturbation to its charge-state because they
24
correspond to charge-separated excited states. Thus the
similarity between the optical absorption spectra of the previ-
36
Fig. 9), consistent with previous reports.
4
ꢀ
ously reported [Ag (SR) ]
and the clusters in this work
4
4
30
What differentiates the electronic structure of the
suggests that the two are identical in the superatom charge-
state, in addition to the stoichiometric formula and size, which
were conrmed by ESI mass spectrometry and other techniques
Ag (MNBA) NCs from the bulk Ag is the absence of the
4
4
30
intensity at the Fermi edge region; the onset of the NCs spec-
trum is around 0.9 eV away from the Fermi edge (the inset in
Fig. 9), which is strong evidence of the semiconducting nature
of the Ag44(MNBA)30 NC lms. The other clearly noticeable
difference is the feature at 9.7 eV observed in the NC's spectra
that is not present in the spectra of bulk Ag. This feature is
tentatively attributed to the organic content of the lm,
although the nature of the valence band structure e.g. the origin
of the spectral features is not well understood at the moment
and is the subject of further investigation.
(vide supra).
The valence band structure of the NCs deposited onto an Au

lm is shown in Fig. 9. The (dashed) black curve corresponds to
Conclusion
A simple, scalable and reproducible protocol for the synthesis of
highly stable, water-soluble and atomically monodisperse
Ag (SR) NCs is described. The clusters can be stored at room
4
4
30
temperature for many months, and can be used to fabricate
lms that retain the unique optical absorption properties of the

dispersed NCs. We anticipate that this development will facili-
tate further studies of the cluster's physical and electronic
properties, and crystallization for full structural determination
through single crystal X-ray diffraction. Moreover, the favour-
able optical properties of Ag (SR) combined with stability
4
4
30
puts the utilization of the clusters in light harvesting applica-
tions within reach. Those topics are the subject of our current
and future investigation.
Fig. 8 UV-Vis spectra of Ag44(MNBA)30 NCs (red curve) in comparison with
Ag44(4-FTP)30 ones (black curve). The inset shows the chemical structures of
MNBA and 4-FTP.
This journal is ª The Royal Society of Chemistry 2013
J. Mater. Chem. A, 2013, 1, 10148–10154 | 10153

View Article Online
Journal of Materials Chemistry A
Paper
1
5 N. Cathcart, P. Mistry, C. Makra, B. Pietrobon, N. Coombs,
M. Jelokhani-Niaraki and V. Kitaev, Langmuir, 2009, 25,
5840–5846.
Conflict of interest
The authors declare no competing nancial interests.
16 S. Kumar, M. D. Bolan and T. P. Bigioni, J. Am. Chem. Soc.,
2
010, 132, 13141–13143.
Acknowledgements
1
7 A. Desireddy, S. Kumar, J. Guo, M. D. Bolan, W. P. Griffith
and T. P. Bigioni, Nanoscale, 2013, 5, 2036–2044.
The authors acknowledge the use of KAUST's Core Labs, the
Canadian Light Source (CLS) synchrotron facility and NSF 18 F. Bertorelle, R. Hamouda, D. Rayane, M. Broyer, R. Antoine,
0
903787 (A.D.). L.A. thanks Dr Maher Fathalla, Dr Alain
P. Dugourd, L. Gell, A. Kulesza, R. Mitric and V. Bonacic-
Koutecky, Nanoscale, 2013, 5, 5637–5643.
LeSimple, and Dr Xiaoyu Cui for helpful discussions and useful
scientic insights.
19 N. Cathcart and V. Kitaev, J. Phys. Chem. C, 2010, 114, 16010–
1
6017.
0 T. U. B. Rao, B. Nataraju and T. Pradeep, J. Am. Chem. Soc.,
010, 132, 16304–16307.
2
References
2
1
2
A. Dass, Nanoscale, 2012, 4, 2260–2263.
M. M. Alvarez, J. T. Khoury, T. G. Schaaff, M. N. Shagullin,
21 H. Yao, M. Saeki and K. Kimura, J. Phys. Chem. C, 2010, 114,
15909–15915.
I. Vezmar and R. L. Whetten, J. Phys. Chem. B, 1997, 101, 22 T. Udaya Bhaskara Rao and T. Pradeep, Angew. Chem.,
706–3712.
Int.Ed., 2010, 49, 3925–3929.
H. Qian, M. Zhu, Z. Wu and R. Jin, Acc. Chem. Res., 2012, 45, 23 K. M. Harkness, Y. Tang, A. Dass, J. Pan, N. Kothalawala,
3
3
4
5
1
470–1479.
V. J. Reddy, D. E. Cliffel, B. Demeler, F. Stellacci and
O. M. Bakr, Nanoscale, 2012, 4, 4269–4274.
24 M. Pelton, Y. Tang, O. M. Bakr and F. Stellacci, J. Am. Chem.
Soc., 2012, 134, 11856–11859.
G. Li and R. Jim, Acc. Chem. Res., 2013, 5, DOI: 10.121/
ar400144e.
H. Qian, Y. Zhu and R. Jin, Proc. Natl. Acad. Sci. U. S. A., 2012,
1
09, 696–700.
25 B. Demeler, http://www.ultrascan.uthscsa.edu.
26 R. P. Carney, J. Y. Kim, H. Qian, R. Jin, H. Mehenni,
F. Stellacci and O. M. Bakr, Nat. Commun., 2011, 2, 335.
27 L. A. Angel, L. T. Majors, A. C. Dharmaratne and A. Dass, ACS
Nano, 2010, 4, 4691–4700.
6
7
A. Dass, J. Am. Chem. Soc., 2011, 133, 19259–19261.
I. Dolamic, S. Knoppe, A. Dass and T. B u¨ rgi, Nat. Commun.,
2012, 3, 798.
8
M. Zhu, C. M. Aikens, M. P. Hendrich, R. Gupta, H. Qian,
G. C. Schatz and R. Jin, J. Am. Chem. Soc., 2009, 131, 2490– 28 T. G. Schaaff, G. Knight, M. N. Shagullin, R. F. Borkman
492.
and R. L. Whetten, J. Phys. Chem. B, 1998, 102, 10643–10646.
P. D. Jadzinsky, G. Calero, C. J. Ackerson, D. A. Bushnell and 29 M.-C. Bourg, A. Badia and R. B. Lennox, J. Phys. Chem. B,
R. D. Kornberg, Science, 2007, 318, 430–433.
2000, 104, 6562–6567.
0 H. Yang, J. Lei, B. Wu, Y. Wang, M. Zhou, A. Xia, L. Zheng 30 O. A. Wong, C. L. Heinecke, A. R. Simone, R. L. Whetten and
and N. Zheng, Chem. Commun., 2013, 49, 300–302.
C. J. Ackerson, Nanoscale, 2012, 4, 4099–4102.
1 H. Yang, Y. Wang and N. Zheng, Nanoscale, 2013, 5, 2674– 31 Z. Hens and J. C. Martins, Chem. Mater., 2013, 25, 1211–1221.
2
9
1
1
1
1
2
677.
32 R. C. Price and R. L. Whetten, J. Am. Chem. Soc., 2005, 127,
13750–13751.
33 J. A. Dale and H. S. Mosher, J. Am. Chem. Soc., 1973, 95, 512–519.
2 Y. Negishi, R. Arai, Y. Niihori and T. Tsukuda, Chem.
Commun., 2011, 47, 5693–5695.
3 I. Chakraborty, A. Govindarajan, J. Erusappan, A. Ghosh, 34 M. Walter, J. Akola, O. Lopez-Acevedo, P. D. Jadzinsky,
T. Pradeep, B. Yoon, R. L. Whetten and U. Landman, Nano
Lett., 2012, 12, 5861–5866.
4 O. M. Bakr, V. Amendola, C. M. Aikens, W. Wenseleers, R. Li,
G. Calero, C. J. Ackerson, R. L. Whetten, H. Gr ¨o nbeck and
H. H ¨a kkinen, Proc. Natl. Acad. Sci. U. S. A., 2008, 105,
9157–9162.
1
L. Dal Negro, G. C. Schatz and F. Stellacci, Angew. Chem., 35 C. M. Aikens, J. Phys. Chem. Lett., 2010, 2, 99–104.
009, 121, 6035–6040.
36 L. Gerenser and R. Baetzold, Surf. Sci., 1980, 99, 259–268.
2
1
0154 | J. Mater. Chem. A, 2013, 1, 10148–10154
This journal is ª The Royal Society of Chemistry 2013