Metal Doping of Au25(SR)18- Clusters: Insights and Hindsights

Article abstract of DOI:10.1021/jacs.9b08228

The study of the structures and properties of atomically precise gold nanoclusters is the object of active research worldwide. Recently, research has been also focusing on the doping of metal nanoclusters through introduction of noble metals, such as plat

Full text of DOI:10.1021/jacs.9b08228

Subscriber access provided by - Access paid by the | UCSB Libraries
Article
25
18-
Metal Doping of Au(SR) Clusters: Insights and Hindsights
Wenwen Fei, Sabrina Antonello, Tiziano Dainese, Alessandro Dolmella,
Manu Lahtinen, Kari Rissanen, Alfonso Venzo, and Flavio Maran
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.9b08228 • Publication Date (Web): 18 Sep 2019
Downloaded from pubs.acs.org on September 19, 2019
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 14
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
−
Metal Doping of Au25(SR)18 Clusters: Insights and Hindsights
Wenwen Fei,† Sabrina Antonello,† Tiziano Dainese,† Alessandro Dolmella,≠ Manu Lahtinen,♯ Kari
9
Rissanen,♯ Alfonso Venzo,‡ and Flavio Maran†,§,*
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
† Department of Chemistry, University of Padova, via Marzolo 1, 35131 Padova, Italy
≠ Department of Pharmaceutical and Pharmacological Sciences, University of Padova, via Marzolo 5, 35131 Padova, Italy
♯ University of Jyvaskyla, Department of Chemistry, P.O. Box 35, 40014 Jyväskylä, Finland
‡ National Research Council, CNR-ICMATE, Department of Chemistry, University of Padova, via Marzolo 1, 35131 Padova
§ Department of Chemistry, University of Connecticut, 55 North Eagleville Road, Storrs, Connecticut 06269, United States
Keywords: atomically precise gold nanoclusters, Au25(SR)18, metal doping, NMR of doped gold nanoclusters,
electrochemistry of doped gold nanoclusters, single crystal X-ray crystallography
E-mail: flavio.maran@unipd.it
Abstract. The study of the structures and properties of atomically precise gold nanoclusters is the object of active research
worldwide. Recently, research has been also focusing on the doping of metal nanoclusters through introduction of noble metals,
such as platinum, and less noble metals, such as cadmium and mercury. Previous studies, which relied extensively on the use of
mass spectrometry and single crystal X-ray crystallography, led to assign the location where each of these foreign-metal atoms go.
Our study provides new insights into this topic and, particularly, compelling evidence about the actual position of the selected metal
atoms M = Pt, Pd, Hg, and Cd in the structure of Au24M(SR)180. To make sure that the results were not dependent on the thiolate,
for SR we used both butanethiolate and phenylethanethiolate. The clusters were prepared according to different literature
procedures that were supposed to lead to different doping positions. Use of NMR spectroscopy and isotope effects, with the support
of mass spectrometry, electrochemistry, and single crystal X-ray crystallography, led us to confirm that noble metals indeed dope
the cluster at its central position, whereas no matter how the doping reaction is conducted and the nature of the ligand, the position
of both Cd and Hg is always on the icosahedron shell, rather than at the central or staple position, as often reported. Our results not
only provide a reassessment of previous conclusions, but also highlight the importance of NMR spectroscopy studies and cast
doubts on drawing conclusions mostly based on single crystal X-ray crystallography.
INTRODUCTION
tetrachloroauric and hexachloroplatinic acids with a given
thiol, followed by sodium borohydride reduction.12-15 The
same procedure was applied to palladium14,16-19 and mercury.15
Cadmium20-22 and mercury20-23 were introduced into preformed
Au25(SR)18−, and studied from several viewpoints. Regarding
cadmium, Wu and co-workers reported that when the metal
source is a salt, Cd(NO3)2, doping occurs on the icosahedron.21
Many thiolate-protected gold nanoclusters, especially those
sufficiently small (typically, less than ca. 144 atoms) to
display electrochemical,1 optical,2 and magnetic3 molecular
properties, can be prepared with atomic precision.4,5 Recently,
research has also been focusing on the selective doping of
metal nanoclusters through introduction of foreign-metal
atoms.6-9 This is a very important area for both fundamental
and applied (e.g., catalysis) purposes. Modification of the
metal composition has been studied for several clusters, but
most research has focused on Au25(SR)18, which is an
atomically precise cluster that has been long considered a
convenient benchmark system for understanding properties
and devising applications of gold nanoclusters.10,11
0
The analysis of the data pertaining to Au24Cd(SC2Ph)18
(SC2Ph = phenylethanethiolate; hereafter, we will indicate the
number of carbon atoms of the alkyl chain simply as Cn)
relied on X-ray crystallography and theoretical calculations of
the experimental UV-vis-NIR spectrum, in comparison with
the
corresponding
mercury
monodoped
cluster
0
Au24Hg(SC2Ph)18 and the pertinent matrix-assisted laser
desorption ionization time-of-flight (MALDI-TOF) mass
spectrometry fragmentation patterns. The latter also was
obtained using a salt as the metal source, Hg(NO3)2.23
Interestingly, the two very similar syntheses led to doping at
different positions: whereas Cd would go on the
icosahedron,21 for Hg the X-ray single-crystal diffraction
results were interpreted to indicate that one of the staple Au
atoms is replaced by Hg. Theoretical simulations of the
experimental UV-vis-NIR spectrum, and the MALDI-TOF
−
Controlled doping of Au25(SR)18 has been carried out with
the noble metals platinum12-16 and palladium,14,16-19 and less
noble metals, such as cadmium20-22 and mercury,15,20-23 also
because of the ease by which monodoping could be achieved
with these metals as opposed to, say, copper and silver.9 Mass
spectrometry and single crystal X-ray crystallography were
extensively employed to interpret the doping results and,
particularly, assign the specific locations where these single
foreign-metal atoms go. The platinum-doped clusters were
prepared by direct synthesis, i.e., by reacting a mixture of
mass-spectrometry,
thermogravimetric
analysis,
and
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 14
electrochemical results provided further support to this
conclusion.23 Further work used this conclusion to understand
what happens when Ag is used to dope a preformed
1
2
3
4
5
6
7
8
0
0
Au24Hg(SC2Ph)18 cluster.24 For both Au24Hg(SC2Ph)18 and
the so-formed trimetallic cluster, the NMR results were taken
as a further indication that Hg atom most probably occupies a
staple position.
Cadmium and mercury were introduced by the Zhu group
−
into preformed Au25(SC2Ph)18 by using a different approach
in which the metal is added to the cluster solution as a thiolate,
Cd(SC2Ph)2 or Hg(SC2Ph)2.20 For Cd, single-crystal X-ray
crystallographic data indicated that doping occurred at the
central position. As to Hg, the cluster was concluded to have
the same structure due to the same valence, NMR spectrum,
and the loss of the same M1Au4 fragment in MALDI-TOF
mass spectrometry. NMR was used only to rule out the
possible presence of the tetraoctylammonium countercation,
9
0/−
Figure 1. (a) Structure of Au25(SR)18
showing the three
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
position types that can be occupied upon metal monodoping:
central (blue), icosahedron (pink), and staple (green). The gold
(yellow) and sulfur (red) atoms are shown, whereas the
carbonaceous part of the ligands is omitted for clarity. (b)
Structure of Au25(SR)18 showing the carbons (gray), hydrogens
(white), and terminal groups (black) for both ligand types of one
of the six staples.
0/−
−
present in the native Au25(SC2Ph)18 solution. As for the
metal-salt studies,21,23,24 the NMR spectrum showed a complex
pattern. The outcome of the stepwise formation of trimetallic
clusters, MAgxAu24-x(SC2Ph)18, was interpreted25 on the basis
of the aforementioned conclusions on the central position
eventually occupied by Cd or Hg.20 Some doped clusters were
prepared with ligands other than phenylethanethiol. In
particular, Thanthirige et al. prepared Au24M(SC6)180 (M = Pt,
Hg) clusters by direct synthesis. Analysis of the MALDI-TOF
mass spectrometry and X-ray photoelectron (XPS) spectra led
to conclude that for both metals the doping occurred at the
central position.15 The preparation and other properties of
Theoretical calculations have been performed to predict or
explain the position of the heteroatom upon monometal
doping, even ahead of substantial experimental work. Earlier
DFT calculations by Jiang and Dai pointed to Cd and Hg as
stable when in the center position.32 On the other hand, other
DFT calculations carried out by Walter and Moseler predicted
that Pd should be more stable when at the center, whereas for
Cd the lowest energy isomer is at the icosahedron, rather than
elsewhere.33 In a recent study, Taylor and Mpourmpakis
used34 their thermodynamic stability model (TSM), which
attributes structure stability to a balance between the chemical
potentials of the metal atoms in the core and the protecting
shell,35 to describe doping effects on nanoclusters. The TSM
predictions were concluded to be in excellent agreement with
experiments. The case of Hg is particularly interesting. As we
saw, Hg has been described as being at the center15,20 or in a
staple position,23 with the latter considered34 as more likely
and in agreement with the TSM. Although the icosahedral
position, Hg(i), resulted close to the 95% prediction interval, a
better proximity of the Hg(s) to the parity line (in a plot
between shell-to-core bond energy and the metal-core
cohesive energy) was seen as providing the first theoretical
rationalization for the experimental observation of the Hg(s)
position in Au24Hg(SR)180. Regarding Cd, which was also
described that, depending of the synthetic method, could
occupy two positions, center20 or icosahedron,21 the TSM
results pointed to Cd(c) doping as being closer to the parity
line than Cd(i) doping; thus, the authors also suggested that
the latter could be potentially transformed into the former
under proper experimental stimulus. Very recently, the Aikens
group used DFT to study the doping process in a few clusters,
including Au24M(SR)18.36 Whereas group X dopants (Pd, Pt)
resulted stable when at the central position, for dopants in
groups XI− XIII (and thus also for Cd and Hg, group XII) the
icosahedral position was found thermodynamically preferable
mainly due to group theory and relativistic effects. As to the
staple position compared to the central position, whereas for
Cd the former has a slightly lower energy, for Hg the results
point to Hg(c) as being quite more stable than Hg(s).
0
Au24M(SC6)18 (M = Pt, Pd) were described by the Lee and
Jiang groups in a previous publication.14 Negishi et al. used
0
dodecanethiol to prepare Au24Pd(SC12)18 and concluded, on
the basis of experiments and density-functional theory (DFT)
calculations that Pd occupies the center of the core.17,18
In a recent review article,8 Zhu and co-workers concluded
that "Doping specific number of heterometal atoms into
specific positions of the nanocluster template is still one of the
most challenging tasks in the nanofield." We could not agree
more. Indeed, now the question is: How can we assign the
specific position where these foreign-metal atoms actually go
to? This is not just a problem per se, but also has far-reaching
consequences because the results described above are
consistently taken as the starting point for other investigations,
whether related to different clusters or applications, as
discussed in several review articles.5,7-9,11,25-28 Here we address
this problem by specifically focusing on the doping with Pt,
Pd, Cd, and Hg atoms to form the corresponding
Au24M(SR)180 clusters. As aforementioned, conclusions on the
specific location of the foreign-metal atom have been drawn
mostly on the basis of the interpretation of single-crystal X-
ray crystallography and MALDI-TOF mass spectrometry data,
sometimes with the support of DFT calculations, and results
from XPS and UV-vis absorption spectroscopy. The structure
of Au25(SR)18, whether in the anionic or neutral form,29-31 is
0 16,19,20,21,23
maintained in Au24M(SR)18
,
and shows that there
are three possible positions for the M atom: center (c),
icosahedron (i), and staples (s) (Figure 1a); whereas there is
only one central atom, the other positions are of 12-fold
equivalency.
With the exception of Pt and Pd, for which there is no doubt
that the direct synthesis yields a cluster monodoped at its
central position, it is thus clear that for Cd and Hg available
experimental data provide a number of opposite conclusions,
possibly related to the specific synthetic method. DFT
ACS Paragon Plus Environment

Page 3 of 14
Journal of the American Chemical Society
calculations have provided hints on this topic, though some
analyses appear to be in contrast at least to some experimental
shown in Figure S2 for Au24Pt(SC4)180, and no contamination
0
from residual Au25(SC4)18− or Au25(SC4)18
.
1
2
3
4
5
6
7
8
0
0
conclusions.
Here
we
use
NMR
spectroscopy,
The NMR behaviors of Au24Pt(SC4)18 and Au24Pd(SC4)18
electrochemistry, MALDI-TOF, and single-crystal X-ray
crystallography of each sample to demonstrate that in several
cases the conclusions reached on the actual position of Cd
and Hg atoms need to be drastically revised. Our study
includes: (i) clusters protected by SC4 (Pt, Pd, Cd, Hg) and
the SC2Ph (Cd, Hg) as the ligands; (ii) direct synthesis (Pt,
Pd) and indirect synthetic methods, that is, metal exchange on
were studied in C6D6, at 1.5-2.1 mM concentration of the
cluster, and the chemical shifts () are referred to
tetramethylsilane; these conditions were the same also for all
1
other clusters studied. Beside the monodimensional H NMR
spectra, the clusters were studied by 1H,1H-homonuclear
correlation spectroscopy (COSY) and 1H,13C-heteronuclear
multiple quantum coherence (HMQC) spectroscopy. The 2D
spectra, whose analysis allowed assigning all resonances, are
provided in Figures S3-S5, whereas Table S1 gathers all
−
0
9
both Au25(SR)18 (Cd, Hg) and Au24Cd(SR)18 (Hg); (iii) for
Cd and Hg we used both the metal salt and the metal thiolate
methods. This study is meant to provide new insights and
perspectives into this general problem, and describe a possible
experimental methodology to understand the actual doping
location. The power of NMR spectroscopy and associated
isotopic effects are especially highlighted. These results call
for a warning about the reliability of conclusions based on
mass-spectrometry fragmentation patterns and, especially, X-
ray crystallography of doped clusters.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
0
chemical shift values. Au24Pt(SC4)18 and Au24Pd(SC4)18 are
diamagnetic, neutral species16 and are thus directly
comparable with the diamagnetic anion [Au25(SC4)18−](n-
Oct4N)+ (where n-Oct stands for norm-octyl), whose NMR
data37,38 are also provided in Table S1.
RESULTS AND DISCUSSION
0
Synthesis. Au24Pt(SC4)18 was prepared by reacting a
solution of HAuCl4 and H2PtCl6 with the given thiol, followed
0
by addition of NaBH4.12 Au24Pt(SC4)18 could be purified
from the main co-product, Au25(SR)18−, according to the
procedure described by Qian et al.,12 in which H2O2 is used to
cause degradation of the undoped cluster through multiple
oxidation processes. For reasons that will be discussed in the
X-ray crystallography section, we used n-butanethiol. The
0
synthesis of Au24Pd(SC4)18 was carried out according to a
very similar protocol, but for the use of Na2PdCl6 in place of
−
H2PtCl6. Two Au25(SR)18 clusters (R = C4, C2Ph), which
were prepared as already described,37-39 were allowed to react
with Cd(NO3)2 or Hg(NO3)2, as described by the Wu
group,21,23 and Cd(SR)2 or Hg(SR)2, as described by the Zhu
group.20 These reactions are described in detail in the
Experimental Section; we found that the same protocol works
well for both C4 and C2Ph. In addition to using Cd(NO3)2 for
making the thiolate, we used CdCl2, and the reactions went
equally well. This check was expedient to then carry out the
−
exchange reactions on Au25(SR)18 clusters with 113Cd(SR)2.
0
Finally, on the two Au24Cd(SR)18 clusters we exchanged Cd
with Hg, as described by the Wu group for R = C2Ph.21 The
clusters were carefully purified, recrystallized, and only
afterward each sample batch was used for the NMR
spectroscopy, UV-vis absorption spectroscopy (Figure 2
shows the SC4 series, whereas Figure S1 shows the SC2Ph
series), MALDI-TOF mass spectroscopy, cyclic voltammetry
(CV), and differential pulse voltammetry (DPV)
measurements. The synthetic methods used only yielded
monodoped clusters. Most of them were also studied by single
crystal X-ray crystallography. To check the quality of the
results further, some of the crystals studied at the University
of Jyväskylä were also studied at the University of Padova.
0
Au24Pt(SC4)18 and Au24Pd(SC4)180. For both platinum
Figure 2. UV-vis absorption spectra of all SC4 samples (0.2 mM,
1 mm cuvette) in CH2Cl2. For the sake of better comparison, the
curves have been shifted vertically. The dashed lines mark the
corresponding zero absorbance.
0
and palladium, the direct synthesis of Au24M(SR)18 (R =
C2Ph, C6, C12) has been consistently described to yield
clusters doped at the center.12-19 For R = C4, we followed the
same synthetic and purification protocol described for
platinum by Qian et al.12 and adapted for palladium by Kwak
et al..14 MALDI-TOF mass spectrometry clearly indicated that
the purified clusters only contain one foreign-metal atom, as
Au25 and monodoped clusters are known to share the same
structural features: a central Au atom, 12 Au atoms forming an
icosahedron, and an external shell composed of six –(SR)in–
ACS Paragon Plus Environment

Journal of the American Chemical Society
Au–(SR)out–Au–(SR)in– double staples. Each staple consists of
Page 4 of 14
1
2
3
4
5
6
7
8
two inner thiolates (in) and one outer thiolate (out) (Figure
1b). The term inner indicates that the two SR groups also bind
to the icosahedron Au atoms, whereas outer indicates that the
SR group is at the outmost position of the double staple. In
[Au25(SC4)18−] (n-Oct4N)+,37,38 the two ligand types have
different  values, well-defined signals (corresponding to the
methylene groups in positions , , and  with respect to
sulfur, and the methyl group in position ), and for the same
resonance the integrals are in the 2:1 ratio expected for the 12
9
0
inner and 6 outer ligands. The spectra of Au24Pt(SC4)18 and
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
Au24Pd(SC4)18 exhibit exactly the same general features, but
for slightly different chemical-shift values (Table S1). Figure
3 shows the comparison between the Pt-doped and the
undoped clusters. This behavior clearly indicates that the
−
symmetry of the ligands of the parent Au25(SC4)18 cluster is
preserved upon monodoping, as also discussed for R = C2Ph
by Qian et al.12 and later by Tian et al.16 For both Pt and Pd,
1
0
the H NMR spectra of Au24M(SC4)18 can thus be taken as
representing the blueprint of the typical "uncomplicated"
NMR behavior expected for a Au24M(SC4)180 cluster doped in
its central position.
Figure 4. 1H NMR spectra of 2.1 mM Au24Hg(SC4)180 in C6D6 at
25 C. The Greek symbols have the usual meaning. The inset
shows an enlarged part of the spectral region (experiment and
simulation) pertaining to 11 -(CH2)in resonances. Integrals refer
to the number of protons.
(Figures S7 and S8); the integrals, carried out in the ranges
indicated, are in a 2:1 ratio, as expected for the inner relative
to the outer resonances. Figure 4 clearly shows that for this
cluster the ligand symmetry is completely removed. Table S2
1
gathers all H and 13C chemical shifts. Perturbations from the
simple pattern exhibited by the undoped cluster (Figure 3a)
are seen for all resonances and are especially evident for the
inner ligands and the protons nearer to the cluster core. It is
noteworthy that signal complexity is observed even for the -
(CH3)out proton resonance, which, being the most distant from
the core, in all gold nanocluster previously investigated3 has
always been the least sensitive to core size, charge, magnetic
state, and environmental effects. The -(CH2)in and -(CH2)out
resonances, which are at 4.07-3.78 and 3.15-3.05 ppm,
respectively, exhibit particularly complex patterns. In
particular, one of the -(CH2)in triplets (with an integral value
corresponding to one ligand) is clearly separated from the
others. It is also worth noticing that this lack of symmetry
does not induce diastereotopic effects in the ligands, as
opposed to what found for achiral ligands in the presence of
interligand interactions and/or when the staple arrangement is
1
Figure 3. H NMR spectra of (a) [n-Oct4N+] [Au25(SC4)18–] and
(b) Au24Pt(SC4)180. The peaks marked with a star pertain to n-
Oct4N+. Both samples were in C6D6 at 25 C.
0
0
chiral (Au38(SR)24 and Au144(SR)600).40-42 Regarding the
Au24Hg(SC4)18
.
The mercury-doped clusters were
complexity of the proton signals for both the inner and outer
ligands (see below), a quite similar behavior is also exhibited
by the corresponding 13C resonances and 13C chemical shift
values (HMQC experiments, Figure S9). In particular, for one
prepared according to three previously published methods (for
−
C2Ph), i.e., by the (i) Au-exchange reaction of Au25(SC4)18
with Hg(SR)2 (Wang et al.)20 and (ii) Hg(NO3)2 (Liao et al.),23
0
and by the (iii) Cd-exchange reaction of Au24Cd(SC4)18 with
1
Hg(NO3)2 (Yao et al.).21 After purification and
recrystallization, all three methods led to obtain very pure
Au24Hg(SC4)180 samples. The UV-vis absorption spectroscopy
and MALDI-TOF spectra are shown in Figures 2 and S6,
respectively.
of -(CH2)in carbons, which corresponds to the isolated H
triplet at 4.055 ppm, the 13C chemical shift value is distinctly
smaller (37.25 ppm) than the similar values shown by the
other 11 ligands (39.01-38.79 ppm). A few differences are
also detected for the -(CH2)in, -(CH2)in, and -(CH3)in 13C
resonances. Small differences are also present in the -
(CH2)out and -(CH2)out resonances, whereas -(CH2)out and -
(CH3)out appear isochronous.
We first consider the Au24Hg(SC4)180 sample prepared from
Hg(SR)2. Figure 4 shows its 1H NMR spectrum and
identification of the signals, as achieved by analysis of the
COSY and total correlation spectroscopy (TOCSY) spectra
Regarding the number of different inner ligands detectable
in the 1H NMR spectrum, we took advantage of the net
separation of the -(CH2)in triplet at 4.055 ppm to simulate the
cumulative signal pertaining to the remaining 11 inner
ACS Paragon Plus Environment

Page 5 of 14
Journal of the American Chemical Society
resonances. By using the intensity of the isolated triplet as the
1H NMR spectrum of the purified, recrystallized cluster
obtained according to the Cd(NO3)2 method. Regardless of the
synthetic procedure, however, the 1H NMR spectra are
1
2
3
4
5
6
7
8
starting point, we generated the convoluted signal by
assigning and optimizing the chemical shifts of the 11 triplets.
The indent in Figure 4 shows the satisfactory outcome of the
simulation. The individual chemical shifts (the number of
isochronous signals is given in parenthesis) are at 3.883 (1),
3.862 (2), 3.834 (1), 3.823 (4), 3.805 (1), 3.798 (1), and 3.794
(1) ppm.
1
identical (Figure S11). As for the Hg case, the H NMR
spectra show that the ligand symmetry is removed. The
various signal types (position along the ligand chain and
ligand type) were attributed through TOCSY analysis. The -
(CH2)in and -(CH2)out resonances at 3.93-3.67 and 3.16-3.03
ppm, respectively, exhibit a complex pattern qualitatively
similar to that of the Hg-doped clusters. Main differences are:
(i) one of the -(CH2)in is separated from the others but is seen
at higher fields (at 3.716 ppm); (ii) the separation of one of the
-(CH2)out resonances (at 3.052 ppm) is more evident than for
the Hg case. The integrals (Figure 5a) of the various inner and
outer resonances are also in a 2:1 ratio.
A very similar set of results was obtained by analysis of the
TOCSY spectrum (Figure S8): 3.883 (1), 3.854 (2), 3.818 (3),
3.808 (3), 3.786 (2) ppm. Despite small differences between
the two methods, these results show that the perturbation
caused by replacing one single Au atom with Hg generates at
least 6 subgroups in the -(CH2)in resonances. Regarding the
-(CH2)out triplets, analysis of the TOCSY spectrum shows
that the complex signal at ca. 3.1 ppm is composed by 6
distinguishable triplets: 3.147 (1), 3.134 (1), 3.129 (1), 3.109
(1), 3.087 (1), and 3.053 (1) ppm. It is noteworthy that the
effect of Hg is so strong that even all other inner and outer
resonance types are affected. The 1H and 13C data are collected
in Table S2.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Overall, these results clearly show that Hg cannot be
located in the center of the cluster, as previously proposed for
R = C2Ph.20 On the other hand, the above NMR data cannot
conclude whether Hg is in one of the icosahedral vertexes or
one of the staples.
The second sample investigated was prepared from
Hg(NO3)2, according to the method described by Liao et al..23
It should be recalled that this method was described to yield
an Hg(s) doped cluster, though starting from a different ligand
1
(R = C2Ph). The H NMR spectrum (Figure S10), associated
2D spectra, and integral values of recrystallized cluster are
identical to those just described for the first sample.
0
In the third synthetic method, we prepared Au24Cd(SC4)18
(thiolate method, as described in the next section) and then
reacted it with Hg(NO3)2 according to the protocol described
for R = C2Ph by the Wu group.21 The reaction proceeded
rapidly (<10 min) and efficiently (90% yield). Once again, the
1H NMR spectrum of the crystalline product turned out to be
identical to those recorded for the other two samples (Figure
S10). The spectrum of this specific sample was also tested for
stability and found to be perfectly reproducible after 4 weeks
at 10 °C.
It is finally noteworthy that all three samples gave the same
fragmentation pattern in MALDI-TOF mass spectrometry
(Figure S6) and identical UV-vis absorption spectra (Figure
2). Their electrochemical behavior will be discussed later.
0
Figure 5. (a) 1H-NMR spectrum of 2.1 mM Au24Cd(SC4)18
prepared from Cd(NO3)2 in C6D6 at 25 °C. Integrals refer to the
number of protons. Graphs (b) and (c) refer to the -(CH2)in and
0
-(CH2)out regions, respectively, for 2.1 mM Au24Cd(SC4)18
(red) and 2.1 mM Au24113Cd(SC4)18 (blue); the latter was
0
As NMR is an extremely sensitive tool to detect even minor
differences in molecular properties and chemical environment
effects,3 the results obtained for the three samples allow us to
conclude that: (i) the specific synthetic approach, including
the indirect method, does not yield clusters doped at different
positions; (ii) mercury does not dope the cluster at the central
position. At this stage, whether the three identical samples
consist of Hg(s) or Hg(i) remains to be understood.
0
prepared using the thiolate obtained from113CdCl2.
1
The H and 13C data (Table S3) show very similar patterns
as observed for the Hg-doped clusters. According to the
TOCSY spectrum, the 12 -(CH2)in triplets are at 3.865 (2),
3.834 (6), 3.826 (2), 3.803 (1), and 3.716 (1) ppm, whereas the
6 -(CH2)out triplets are at 3.126 (1), 3.122 (1), 3.108 (3), and
3.052 (1) ppm. Differences are also seen along the ligand
chain for both inner and outer ligands. Regarding 13C, some
differences are seen for -(CH2)in (whereas 11 ligands are at
38.2  0.2 ppm, the isolated ligand in at 34.15 ppm), -
(CH2)in, -(CH3)in, -(CH2)out, and -(CH2)out; for (CH2)in, -
Au24Cd(SC4)18
.
The cadmium-doped clusters were
prepared in three ways. The first two methods are based on the
use of (i) Cd(SR)2, obtained upon reaction of Cd(NO3)2 with
the thiol (Wang et al.),20 or (ii) Cd(NO3)2 (Yao et al.).21. The
third approach consists in using CdCl2 or 113CdCl2 (instead of
Cd(NO3)2) to make the thiolate. Figure 5a shows the typical
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 14
(CH2)out, and -(CH3)out, differences are undetectable or within
experimental error.
1
2
3
4
5
6
7
8
9
We also note that the NMR pattern of the Cd-doped clusters
does not show any diastereotopic effect, i.e., the protons of
each CH2 in each ligand type and ligand subgroup are
equivalent (as already noted for the Hg-doped cluster). It is
finally worth stressing that all three samples gave the same
fragmentation pattern in MALDI-TOF mass spectrometry
(Figure S12) and identical UV-vis absorption spectra (Figure
2). The electrochemical behavior will be discussed later.
0
To conclude, analysis of the three Au24Cd(SC4)18 samples
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
shows that: (i) independently of the synthetic method doping
always occurs at the same position and (ii) this position is not
at the center of the cluster.
The question now is: is Cd on the icosahedron, as inferred
by Wu and co-workers for SC2Ph,21 or in one of the staples?
To address this problem, we resorted to carry out the Cd(SR)2
synthesis by starting from 113CdCl2. Whereas the natural
abundance of 113Cd in Cd samples is 12.23%, enriched 113Cd
samples contain 95% of this spin 1/2 isotope. A point-by-point
Figure 6. 1H NMR spectra of 2.1 mM Au24113Cd(SC4)180 (a,c)
comparison
between the 1H NMR spectra of
Au24113Cd(SC4)18 and Au24Cd(SC4)18 (both obtained from
the thiolate method) shows (Figure S13) that the effect of the
isotopic enrichment is observed only for the signal at 3.716
ppm, while the rest of the spectrum is completely unchanged.
Figure 5b shows a detail of the only change detected. In
particular, the small bumps around the isolated -(CH2)in
0
0
and 2.1 mM Au24Cd(SC4)18 (b,d) focusing on the -(CH2)in
(a,b) and -(CH2)out (c,d) regions. The spectra are shown
before (blue traces) and after 1H-1H homodecoupling (red
traces). C6D6, 25 °C.
0
triplet of Au24Cd(SC4)18 are significantly enhanced in
Au24113Cd(SC4)180, in agreement with the ca. 8-fold isotopic
enrichment. Conversely, the isolated and the other
5
convoluted -(CH2)out triplets are identical (Figure 5c).
1H-1H homodecoupling experiments were carried out by
applying
a standard pulse sequence. Decoupling was
performed at the frequency of the -(CH2)in signal (2.173
ppm) that exhibits a scalar correlation with the isolated -
(CH2)in signal at 3.716 ppm. Figure 6 shows the main details
0
of the effects observed for the Au24113Cd(SC4)18 and
0
1
Au24Cd(SC4)18 samples. As expected on the basis of the H-
113Cd coupling, whereas the enriched cluster shows a doublet
(Figures 6a), the latter yields a singlet (Figure 6b), though
accompanied by traces of a doublet (due to the presence of
113Cd and 111Cd, which is another spin 1/2 isotope with a
natural abundance of 12.80%). The doublet in Figure 6a
allows determining a 3J(1H-113Cd) coupling constant of
0
14.3(0.1) Hz. For both Au24113Cd(SC4)180 and Au24Cd(SC4)18
,
the same pulse sequence applied to the isolated -(CH2)out
signal (1.719 ppm), which correlates with the corresponding
isolated -(CH2)out signal at 3.052 ppm, transforms the latter
into a sharp singlet, as could be anticipated on the basis of the
uncomplicated shape of this triplet (Figures 6c,d). The slightly
different position of the peaks in the decoupled spectra is due
to the Bloch–Siegert shift, which causes resonances to move
away from the decoupling frequency.
Figure 7. (a) Models of the possible positions occupied by Cd.
3
(b) Karplus-like correlation of the J(1H-113Cd) coupling constant
Figure 7a shows two models of the possible positions
occupied by Cd. For Cd(s), the bond sequence from Cd to the
-(CH2)in and -(CH2)out protons is the same, H-C-S-Cd,
whereas for Cd(i) the distance from the -(CH2)out protons is
larger by two bonds (H-C-S-Au-S-Cd bond sequence) than for
the -(CH2)in protons. It is thus conceivable that for Cd(i)
doping only the -(CH2)in protons are affected, as the
experiments indeed indicate. On the other hand, if the
as a function of the dihedral angle χ2. The Newman projection of
the investigated bond sequence illustrates the relationship
between the χ2 and  dihedral angles. The areas highlighted in
yellow show the regions where the average 3J(1H-113Cd) coupling
constant values are compatible with the experimentally
determined J value.
ACS Paragon Plus Environment

Page 7 of 14
Journal of the American Chemical Society
exchange yields Cd(s), 113Cd would have affected both the -
(CH2)in and -(CH2)out protons with equal probability, which
is in contrast to the experimental outcome.
been adopted by many research groups as sort of a reference
1
2
3
4
5
6
7
8
ligand. We thus studied the Cd- and Hg-doping of
[Au25(SC2Ph)18−](n-Oct4N)+ according to the same sequence
of reactions and tests already described for the SC4 ligand.
To gain more quantitative insights into this aspect, we
propose the use of a Karplus-type correlation, which was
originally developed to describe the dihedral angle
0
Au24Cd(SC2Ph)18 was prepared according to the
Cd(SC2Ph)2,20 Cd(NO3)2,21 and the CdCl2 (or 113CdCl2) -
thiolate methods. As for the SC4 ligand, the corresponding
MALDI-TOF (Figure S14) and UV-vis absorption
1
dependence of three bond H-1H coupling.43 In 1994, Vašák
and co-workers could demonstrate that a Karplus-type
correlation describes nicely also the dihedral angle
dependence of the three bond 113Cd-1H coupling, as obtained
from HMQC data for Cd-substituted metalloproteins in
comparison with the crystal structure data.44,45 The correlation
was observed for the cysteine H-C-S-Cd dihedral angle. The
same group could previously demonstrate that the Cd-
derivative is isostructural with the native protein.46 Vašák and
co-workers concluded that although heteronuclear couplings
involving heavy nuclei generally depend on orbital angular
momentum, electron-nucleus dipole-dipole interaction, and
1
spectroscopy spectra (Figure S1) show no differences. The H
0
NMR spectra of the Au24Cd(SC2Ph)18 sample obtained from
9
Cd(SC2Ph)2 (blue) and Au24113Cd(SC2Ph)18 (red) are shown in
Figure 8. The perfect overlap of the spectra in Figure 8 (but
for the feature magnified in the inset, as discussed below) and
Figure S16, which pertain to the three samples, confirms that
preparing this doped cluster with the thiolate,20 the salt,21 or
the CdCl2 - thiolate method produces the very same result. A
previously reported spectrum (CD2Cl2, Cd(SR)2 synthesis)
shows similar features (Figure S7 in reference 20), whereas no
NMR data were provided in the other report on Cd doping.21
10
11
12
13
14
15
16
Fermi
contact
contributions,
for
Cd-substituted
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
metalloproteins the dihedral angle is the principal determinant
of the Fermi contact term and the dominant variable. Figure
7b, which is adapted from the original work,44,45,47 illustrates
3
the dependence of the J(1H-113Cd) coupling constant on the
3
dihedral angle , J(1H-113Cd) = 36 (cos2) -13 (cos) +1,
with r2 = 98.7% and confidence limits of ca. 10 Hz;
tetrahedral geometry around Cβ is assumed. More precisely,
Figure 7b shows the correlation as a function of 2, which
refers to the dihedral angle with respect to the  carbon atom.
The inset to Figure 7b shows the Newman projection of the
bond sequence and defines the relationship between the
dihedral angles  and 2, where  relates to Ha (one of the two
-(CH2)in protons). The correlation shows that a coupling
constant should be detected no matter the magnitude of  or
2, as at least one of the two protons (Ha and Hb) always
provides a finite coupling value. Regarding our Cd-doped
clusters, the Karplus-type correlation thus confirms that for
the hypothetical Cd(s) some coupling with the -(CH2)out
protons should be observed as well as with the -(CH2)in
protons. This is not seen.
Figure 8. Full overlap of the 1H NMR spectra of 2.2 mM
0
0
Au24Cd(SC2Ph)18 (blue) and 2.2 mM Au24113Cd(SC2Ph)18 (red)
in C6D6 at 25 °C (region of the aliphatic C-H signals). The inset
highlights the effect of isotopic enrichment on the isolated -
(CH2)in resonance. The star marks a solvent impurity (methanol).
Let us now focus on the only resonance -(CH2)in affected
by the presence of 113Cd. Figure 7b shows that very few 2
regions (in yellow; for symmetry, only the range from 0 to
3
180° needs to be considered), which determine the J(1H-
The spectrum shows that the ligand symmetry is disrupted,
which, once again, is inconsistent with Cd(c) doping.20 The
COSY spectrum (Figure S15) allows attributing (Table S4)
the complex signal at 4.0-3.8 ppm (integral corresponding to
11 ligands) to 22 -(CH2)in protons, the slightly distorted
triplet at 3.722 ppm to the twelfth -(CH2)in ligand, and the
multiplet at 3.43-3.21 ppm to the 24 -(CH2)in protons. The -
(CH2)in signal at 3.722 ppm correlates with the -(CH2)in
triplet at 3.287 ppm. The resonances corresponding to the 6
outer ligands appear as a complex multiplet of -(CH2)out
resonances centered at 3.17 (10 protons) followed by one
additional triplet at 3.031 ppm (2 protons), and a series of
largely overlapped -(CH2)out triplets at 3.0-2.9 ppm (10
protons, five ligands) followed by one additional triplet (2
protons) at 2.820 ppm. The -(CH2)out and -(CH2)out triplets
at 3.031 and 2.820 ppm correlate and thus belong to the same
ligand. Careful analysis of the COSY spectrum, which was
acquired during a particularly long time frame, allows
distinguishing 12 nonisochronous inner ligands and 6 outer
3
113Cd) coupling constant values, provide average J(1H-113Cd)
coupling constant values compatible with the experimentally
determined J value of 14.3 Hz, at least within a prudential
uncertainty of ca. ±3 Hz: 0-13° (J = 16.5 ÷ 17.2 Hz) and 128-
143° (J = 17.2 ÷ 11.3 Hz). For steric reasons, however, the
latter is the only plausible region. The reliability of this
0
conclusion will be addressed for Au24Cd(SC2Ph)18
.
0
Au24Cd(SC2Ph)18 and Au24Hg(SC2Ph)180. The results
and conclusions so far reached regard the butanethiolate-
protected clusters and thus the question arises as to whether
they are extendable to other ligands, also considering that the
seminal works carried out by the groups of Zhu and Wu on
Cd- and Hg-doping focused on the phenylethanethiolate
ligand. Indeed, since Donkers et al. described the first
synthesis
and
isolation
of
Au25
protected
by
phenylethanethiolate ligands48 (Note: this cluster was
originally believed to be Au38(SC2Ph)24), followed some years
later by the actual crystallographic structure determination of
[Au25(SC2Ph)18−](n-Oct4N)+,29,30 phenylethanethiolate has
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 14
0
1
ligands (Table S4), which confirms on a quantitative basis the
profound effect of Cd-doping on the ligand symmetry.
Regarding the 13C resonances, the values are pretty much
isochronous (for the same position and ligand type), but for
small differences for the isolated inner and outer signals.
The three Au24Hg(SC2Ph)18 samples exhibit identical H
NMR spectra, even from the viewpoint of minor features
(Figure S19 allows appreciating the perfect correspondence of
the three spectra). The typical 1H NMR pattern of
Au24Hg(SC2Ph)180 is exemplified in Figure 9a, which pertains
to the sample obtained upon metal exchange on a preformed
1
2
3
4
5
6
7
8
The observation of an isolated triplet for both the inner and
outer ligands is compatible with both Cd(i) and, intuitively,
even more for Cd(s), as in this case both the -(CH2)out and -
(CH2)out resonances are significantly affected. Can this be
taken as the proof that the staple position is preferable? To
clarify the position of the Cd atom, we followed the same
procedure used for SC4. First, we checked that the reaction
with CdCl2 proceeds smoothly and then used 113CdCl2 to
prepare Au24113Cd(SC2Ph)180. Figure 8 shows that only the
isolated -(CH2)in is affected by 113Cd, while the rest of the 1H
NMR spectrum is perfectly superimposable to that of
Au24Cd(SC2Ph)180. The inset of Figure 8 shows the detail of
this effect. Decoupling was carried out for both
0
Au24Cd(SC2Ph)18 cluster. Please note that whereas the
Hg(NO3)2 synthesis is supposed to yield Hg(s),23 for the
0
double exchange we purposely used Au24Cd(SC2Ph)18
obtained using Cd(SR)2, as this sample is that supposed to
produce Cd(c).20
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
Au24113Cd(SC2Ph)18 and Au24Cd(SC2Ph)180, and the effects
on both -(CH2)in and -(CH2)out were tested. As expected on
1
the basis of the H NMR spectra, the only effect is on the
isolated -(CH2)in resonance (Figure S17). Not only the effect
is qualitatively the same as described for SC4, but also the
3J(1H-113Cd) coupling constant is quite similar, 13.6(0.2) Hz.
As observed for Au24Cd(SC4)180 (Figure 6), a trace of the -
(CH2)in doublet obtained upon decoupling could be detected
0
also in the Au24Cd(SC2Ph)18 sample. These results confirm
that the Cd dopant is on the icosahedron, as demonstrated
above for the C4 cluster and originally suggested by Wu and
co-workers for the C2Ph cluster.21
According to the correlation between the 3J(1H-113Cd)
0
Figure 9. 1H NMR spectrum of 2.2 mM Au24Hg(SC2Ph)18
coupling constant and the H-C-S-Cd dihedral angle,43,44 the
(aliphatic C-H signals) obtained from Hg(SC2Ph)2 (blue trace) or
after exchange on a preformed Au24Cd(SC2Ph)18 cluster (red
trace). The integrals and ligand types are indicated. The asterisk
marks a solvent impurity (methanol).
0
virtually
identical
coupling
constants
determined
experimentally for the SC4 and SC2Ph ligands point to very
similar H-C-S-Cd dihedral angles. As described later, we
0
could obtain the structures of both Au24Cd(SC4)18 and
The 1H NMR spectrum exhibits four groups of peaks.
Assignments were carried out through COSY measurements
(Figure S20), and the corresponding 1H and 13C chemical
shifts are provided in Table S5. As for Au24Cd(SC2Ph)180, the
inner resonances -(CH2)in and -(CH2)in are clearly separated
from the two outer resonances -(CH2)out and -(CH2)out. For
both -(CH2)in and -(CH2)out, one of the triplets is well
separated from the others. As for SC4, the isolated -(CH2)in
signal is downfield with respect to the group of the remaining
11 ligands, whereas it is upfield for the outer ligands. The
COSY analysis of the -(CH2)in and -(CH2)out resonances
allowed estimating the presence of each of the 12 and 6
ligands, respectively. Nonisochronous signals are also
detected for the corresponding -(CH2)in and -(CH2)out
resonances. As to 13C, the signals are isochronous, with a few
exceptions. Once again, the number of clearly distinguishable
resonances and isochronous signals witnesses the significant
loss of symmetry undergone upon Hg-doping. It is worth
noticing that the 1H NMR spectra obtained for
0
Au24Cd(SC2Ph)18 by single-crystal X-ray crystallography.
Whereas the former is affected by intercluster interactions, the
0
latter refers to unbounded clusters. Au24Cd(SC2Ph)18 thus
provides an ideal case to test on quantitative grounds the
validity of the 3J(1H-113Cd) coupling constant correlation,
which was originally described for metalloproteins, also for
gold nanoclusters. According to the correlation, the
experimentally determined J value corresponds (within ca. 3
Hz) to plausible angle 2 values of 130 ÷ 145° (J = 16.6 ÷
10.6 Hz). In the structure of Au24Cd(SC2Ph)180, we find that
the average Cβ-Cα-S-Cd dihedral angle 2 is 149°; similar
values can be obtained from the structures published by Wang
et al., 150°,20 and Yao et al., 133°.21 These figures yield an
average 2 of 144°, which is indeed in excellent agreement
with the estimated range, also considering the usual limits of
comparing solid- to solution-phase results.
0
Au24Hg(SC2Ph)18 was also prepared in three ways: (i)
−
20
metal exchange on Au25(SC2Ph)18 with Hg(SC2Ph)2 and
Hg(NO3)2,23
and metal exchange on preformed
0
Au24Hg(SC2Ph)18 (Figure S16 in reference 24 and Figure S7
Au24Cd(SC2Ph)18 with Hg(NO3)2.21 After purification and
recrystallization, the three Au24Hg(SC2Ph)18 samples were
characterized by UV-vis absorption spectroscopy (Figure S1),
electrochemistry (see next section), and MALDI-TOF mass
spectrometry, which gave the same fragmentation pattern
(Figure S18).
0
in reference 20) show very similar features (our data were
obtained in C6D6, whereas those published data pertain to
CD2Cl2), though they were not specifically discussed. Overall,
the NMR analysis shows that the Hg atom cannot be at the
central position of the cluster, as previously hypothesized.20
ACS Paragon Plus Environment

Page 9 of 14
Journal of the American Chemical Society
As to its actual position, we argued that the fast Hg
exchange on a preformed Au24Cd(SC2Ph)18 cluster occurs on
the icosahedron as well, i.e., directly on the site occupied by
that the MALDI-TOF mass spectra of the samples obtained
0
1
2
3
4
5
6
7
8
for each doped clusters are identical (Figures S6, S12, S14,
and S18). As either foreign-metal atom is on the icosahedron,
differences in the fragmentation patterns observed between the
Hg and Cd doped clusters should not to be taken as indicating
a different doping position.23 Rather, they just reflect the
effect of the specific doping element, as also supported by the
similar fragmentation patterns exhibited for the same doping
metal by the SC4 and SC2Ph protected clusters.
Cd(i), rather than involving
a
complicate molecular
rearrangements where first Hg exchanges Cd(i) and then
switches position with the nearby Au(s) atom. To gain insights
into this problem, we applied the same decoupling sequence
used for the Cd-doped clusters. The goal was to detect a
3
possible J(1H-199Hg) coupling by relying on the fact that
199Hg has a natural abundance of 16.94% and is a spin 1/2
isotope. Assuming that a Karplus-type correlation is valid also
for the three-bond system H-C-S-Hg, one would expect to see
Electrochemistry of Au24M(SR)18. The electrochemical
measurements were carried out in dichloromethane (DCM)
containing 0.1 M tetrabutylammonium hexafluorophosphate
(TBAH), using a glassy carbon (GC) microdisk electrode.
Figure 11a compares the DPV behavior of Au24M(SC4)18 for
M = Au, Pt, Cd, and Hg. Figure 11b shows the DPV behavior
of the Au24M(SC2Ph)18 samples (M = Au, Cd, Hg). As
expected, for both the Hg- and Cd-doped clusters the various
samples exhibit exactly the same DPV pattern and formal
potential (E°) values. This is exemplified for both ligands in
Figure S22, which shows the DPVs of the Hg samples
obtained with the Hg(NO3)2, thiolate, and Cd exchange
methods.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
some J(1H-199Hg) coupling only for -(CH2)in or both -
(CH2)in and -(CH2)out for Hg(i) and Hg(s), respectively.
Radiating the corresponding -(CH2)in signal (assessed via
TOCSY) transforms the -(CH2)in signal into a singlet
accompanied by a doublet (Figure S21a) that allows
calculating a 3J(1H-199Hg) coupling constant of 36 Hz.
Conversely, radiating the -(CH2)out signal transforms the
corresponding isolated -(CH2)out signal into an
uncomplicated singlet (Figure S21b). We also applied the
same decoupling analysis to the isolated -(CH2)in signal of
Au24Hg(SC4)180, obtained the same outcome, and calculated
the very similar value of 37 Hz (Figure 10), whereas no effect
was detected for the isolated -(CH2)out resonance. Although
to the best of our knowledge a Karplus-like dependence has
never been observed for 199Hg, it is conceivable that a periodic
0
0
The DPVs of Au24Hg(SC4)18 and Au24Cd(SC4)18 are
qualitatively similar to that of Au25(SC4)18−. The doped
clusters undergo two successive one-electron oxidations (E°1
and E°2) at 0.364 and 0.684 V, Au24Hg(SC4)180, and 0.332 and
0.636 V, Au24Cd(SC4)180. In the timescale of voltammetry
experiments, both processes are reversible. Further oxidation
processes are detectable at more positive potentials, though
with formation of chemically labile species. The E° for the
first peak of these doped clusters is more positive than that of
3
dependence such as that found for J(1H-113Cd)44,45 should be
3
qualitatively valid also for of the J(1H-199Hg) coupling. The
virtually identical values determined for C2Ph and C4 would
thus point to very similar average dihedral angles, as
determined for the Cd-doped clusters. Most important, these
results provide compelling evidence that Hg-doping, whether
−
Au25(SC4)18 (E°1 = −0.188, E°2 = 0.139 V, respectively)49 by
0.552 and 0.520 V, respectively. For Au24Hg(SC2Ph)180 (E°1 =
0.451, E°2 = 0.703 V) and Au24Cd(SC2Ph)180 (E°1 = 0.430, E°2
= 0.668 V) similar considerations apply. With respect to
−
performed directly on Au25(SR)18 or indirectly on
Au24Cd(SR)180, consistently yields Hg(i), rather than Hg(s).23
−
Au24(SC2Ph)18 (E°1 = −0.077, E°2 = 0.226 V),50 the positive
shifts of E°1 are 0.528 and 0.507 V, respectively. For both Hg
and Cd, this remarkable positive shift was already
observed.23,51
Regarding the first reduction peak, which for
0
0
Au24Hg(SC4)18 and Au24Cd(SC4)18 occurs at −1.23 and
−1.39 V, respectively, the formation of the anion is chemically
irreversible. For the latter, increasing the CV potential scan
rate (v) allows to detect reversibility, and therefore, determine
an E° value of -1.38 V. We described this procedure in detail
for a series of Au25(SR)18 clusters.50,52,53 The electrochemical
0
gap of Au24Cd(SC4)18 can thus be calculated from the E°
difference between the +1/0 and 0/-1 redox couples. The
corresponding HOMO-LUMO gap can then be estimated by
subtracting the charging-energy contribution, obtained from
the E° difference between the +2/+1 and +1/0 states.54 The
value so-obtained, 1.41 eV, is in very good agreement with the
HOMO-LUMO gap of 1.37 eV estimated from the onset of
0
optical absorption (Figure 2). For Au24Cd(SC2Ph)18 we
0
Figure 10. 1H NMR spectra of 2.1 mM Au24Hg(SC4)18
,
observed the same behavior, and this allows calculating an E°
value of −1.26 V for the 0/−1 redox couple. The HOMO-
LUMO gap is thus estimated to be 1.46 eV, to be compared
with that obtained from the optical spectrum (Figure S1), 1.41
eV and the value of 1.4(0.1) eV obtained by time-resolved
spectroscopic analysis.22
focusing on the isolated -(CH2)in region before (blue) and after
(red) 1H-1H homodecoupling at the frequency of the
corresponding -(CH2)in signal. The two spectra are vertically
shifted for clarity and the arrows mark the doublet. C6D6, 25 °C.
To conclude, NMR demonstrates that both Cd and Hg are
exchanged on the icosahedron, no matter the synthetic method
employed or the nature of the ligand. It should be also noted
For both SC4 and SC2Ph ligands, the analysis of the
reduction of the Hg-doped clusters is more complicated
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 14
because the voltammetric peak exhibits features that suggest
interaction with the electrode surface. Furthermore, for both
Hg-doped clusters the peak is irreversible also at high v values
(up to 50 V s-1). However, at low temperature (−45 °C) and
high v some reversibility is detectable, which allows
estimating E°. By comparing this result with the
corresponding E° determined at the same temperature for
respectively; these gaps are thus slightly smaller than those
−
1
2
3
4
5
6
7
8
electrochemically determined for Au25(SC4)18
and
Au25(SC2Ph)18−, 1.30 and 1.34 eV, respectively.49,55 A very
recent time-resolved spectroscopy analysis led for
Au24Hg(SC2Ph)180 and Au25(SC2Ph)18− to the similar values of
1.2(0.1) and 1.3(0.1) eV, respectively.22 Very recent
calculations provided a similar decrease in the HOMO-LUMO
gap energy on going from Cd to Hg, as well as valuable
insights into the electronic effects introduced by dopants.36
0
Au25(SC2Ph)18−, the E° values of the two Au24Hg(SR)18
clusters at 25 °C could be estimated. Calculation of the
HOMO-LUMO gap yields 1.28 (SC4) and 1.29 (SC2Ph),
0
The DPV of Au24Pt(SC4)18 clearly points to a different
9
orbital-energy distribution. It shows two pairs of peaks
corresponding to the formation of the mono- and dication on
the positive-going scan (E° of 0.475 and 0. 853 V), and the
mono- and dianion on the negative-going scan (E° of -0.287
and -0.622 V). Each of these charge states is chemically
stable. For this cluster, a HOMO-LUMO gap of 0.384 eV can
be estimated from the electrochemical data (for the charging
energy correction, we used the potential difference calculated
for the two oxidation peaks, 0.378 V) is much lower than for
the other clusters investigated. A previous electrochemical
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0
analysis carried out on Au24Pt(SC6)18 yielded the similar
value of 0.34 eV.14 With SC4, however, we fail to detect the
−
large potential difference reported for the E°1 of Au25(SC6)18
0 14
and the first reduction peak of Au24Pt(SC6)18
.
Overall, some of our electrochemical data essentially
confirm previous electrochemical conclusions on the position
occupied by Pt upon Au25-doping14 and the effect of Hg-
doping.23 Most important, however, they provide further
compelling evidence that Hg- and Cd-doping always occur on
the same metal site, no matter the ligand and how metal
exchange is carried out.
Single Crystal X-Ray Crystallography. We could solve
the structure of most of the clusters, sometimes also as the
result of different syntheses and in two laboratories, as
0
0
specified: Au24Pt(SC4)18 (Padova), Au24Hg(SC4)18 (from
0
Hg(SC4)2, Jyväskylä), Au24Cd(SC4)18 (from Cd(NO3)2,
Jyväskylä), Au24Cd(SC4)18 (from Cd(SC4)2, Jyväskylä),
Au24Hg(SC2Ph)18 (from Hg(NO3)2, Padova and Jyväskylä),
Au24Hg(SC2Ph)18 (from Hg(NO3)2 + Au24Cd(SC2Ph)18
Padova and Jyväskylä), and Au24Cd(SC2Ph)18 (from
Cd(SC4)2, Jyväskylä). Here we will focus on the most salient
aspects, whereas full discussion on these results is provided in
the Supporting Information.
0
0
0
0
,
0
0
For both Au24Hg(SC4)18 and Au24Cd(SC4)180, the structure
0
shows the very same features discovered for Au25(SC4)18
0 39
(this cluster is a neutral radical)37 and Au25(SC5)18
:
(i) the
clusters form linear polymers of interconnected clusters; (ii)
the connecting staples form S-Au-Au-S dihedral angles of
nearly 90° (for both Hg- and Cd-doping, 81-85°); (iii) the
neighboring clusters are connected via aurophilic Au-Au
bonds. Formation of the polymers is thus granted by a twist-
and-lock mechanism37 in which the orientation of the alkyl
chains and their van der Waals interaction opens up two
opposite sides of the Au-S-Au staples and favor a closer
approach between neighboring clusters, thereby causing
formation of an intercluster Au-Au aurophilic bond. In the
doped clusters, this bond has a similar length, 3.09 Å (Hg-
Figure 11. Comparison between the DPV curves for (top to
0
bottom): (a) Au25(SC4)18−, Au24Pt(SC4)180, Au24Cd(SC4)18
and Au24Hg(SC4)180; (b) Au25(SC2Ph)18−, Au24Cd(SC2Ph)18
,
,
0
doping) and 3.10 Å (Cd-doping), as found in Au25(SC4)18
0
(3.15 Å)37 and Au25(SC5)18 (2.98 ÷ 3.03 Å).39 These results
0
and Au24Hg(SC2Ph)180. Glassy-carbon electrode, DCM/0.1 M
TBAH, 25 °C.
thus point to the importance of the alkanethiolate ligand, and
show that formation of the intercluster aurophilic bond is
possible regardless of the magnetic state of the cluster: in fact,
ACS Paragon Plus Environment

Page 11 of 14
Journal of the American Chemical Society
as opposed to Au25(SC4)180, both the Hg- and Cd-doped
1
2
3
4
5
6
7
8
clusters are diamagnetic, as evinced from the NMR results and
0
previously shown for Au24Hg(SC2Ph)18 by electron
paramagnetic resonance.23
With that being said, we note that Au24Pt(SC4)180, which
was purposely prepared with the SC4 ligand, does not form
polymers. Its structure does not show any rotation of the
staples, which remains virtually parallel, and shows a
relatively large minimum intercluster Au-Au distance of 3.88
0
Å. This is as previously observed for, say, Au25(SC2)18
,
9
which shows a minimum intercluster Au-Au distance of 4.12
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
0 38
Å,54 and Au25(SC3)18
.
Overall, this may suggest that
electronic factors may also play a role in determining the
Figure 12. Overlap of the structures of the Hg- and Cd-doped
clusters with that of Au25(SC2Ph)18
0 39 The Au/M-S core units are
0
different behavior observed for Au24Pt(SC4)18
.
.
shown on the left-hand side (ligands removed), whereas the full
clusters are shown on the right-hand side (metal core units are
faded, for clarity) to evidence full overlapping of the ligands
(mixed colors). The color codes are: red = Hg-doped cluster (from
Hg(NO3)2), blue = Hg-doped cluster (from Hg(SC2Ph)2), green =
Cd-doped cluster, and yellow = Au25(SC2Ph)18.
Crystallographic analysis alone is not distinctive enough for
determining the positions of these doping metals with reliable
accuracy. This is due to the very small electron-density
differences between Au and the Pt, Hg, and Cd metals,
especially for Pt and Hg that differ from Au by only one
electron. Thus, an Au site substituted by Hg or Pt should show
an electron density higher or lower than that of Au by only
1.2%, respectively. This figure will be significantly lowered if
the doping metal is disordered over two or more locations
(~0.1% difference when all 12 icosahedral sites are partially
but evenly occupied; even less if distribution also involves the
staples) and/or if the quality of the crystal is less than ideal.
This implies that for these clusters small differences in
electron densities cannot be determined reliably even with the
highest quality data obtained by the modern in house
diffractometers. However, thorough analysis of the structure
of the Cd doped clusters (the electron-density difference
between Cd and Au is ~60%) using several data sets with the
highest possible data quality (data redundancy of 5) allowed
us to refine the structure of the Cd-doped cluster quite
satisfactorily. Consistently with the NMR analysis, refinement
indicated that Cd is most likely disorderly located on the
icosahedral sites instead of the center or staples. For the Hg
CONCLUSIONS
This study was meant to obtain insights into the monodoping
−
of Au25(SR)18 clusters with foreign metal atoms. Accurate
selection of chemicals, NMR analysis, electrochemical data,
and critical analysis of crystallographic data allowed us to
highlight hindsights into the challenge of understanding where
the foreign-metal atoms are eventually located in the cluster
structure, and how to characterize these quite elusive
nanosystems. Using the NMR results obtained for
0
0
Au24Pt(SC4)18 and Au24Pd(SC4)18 as a reference for the
behavior expected when the cluster is doped in its central
position, we show that Cd- and Hg-doping does not occur at
the central position.20 We also show that the Cd-doping mode
is not different from that of Hg-doping, as opposed to what
previously concluded.21,23 Rather, we find that both Cd- and
Hg-doping occurs in one of the icosahedral positions,
independently of the specific ligand. Equally important, we
demonstrate that the metal-exchange doping methods so-far
developed always yield the very same species. Besides being
important from a fundamental viewpoint, these results are also
liable to impact applications of doped clusters, e.g., in
catalysis, which is a promising growing area of research for
atomically precise metal clusters.56 This is because proper
understanding of the doping site affects the analysis of the
catalytic mechanism.51 Finally, we provide a warning about
reaching conclusions on the doping site on the basis of
different fragmentation patterns in mass spectra, and
especially, single-crystal X-ray crystallography results. We
also hope that these new insights will be useful for
theoreticians as a sound experimental basis to refine their
calculation models.
0
and Pt doped Au24M(SC4)18 clusters, on the other hand, the
electron-density difference is just too small to draw similar
conclusions.
0
Regarding the SC2Ph-protected clusters, Au24Hg(SC2Ph)18
and Au24Cd(SC2Ph)180 show exactly the same structure, where
the orientation of the ligands with respect to plane of the
staple is always of the up-down-up type (Figure 12). This is,
therefore, identical to the ligand orientation seen in
0 39
Au25(SC2Ph)18
,
though different from that observed in
Au25(SC2Ph)18−, which is always of the up-down-down
type.31,32 Finally, we checked the two structures of
0
Au24Hg(SC2Ph)18 obtained upon metal exchange in either
−
Au25(SC2Ph)18 or Au24Cd(SC2Ph)180, and found they are
identical. Further discussion on the SC4- and SC2Ph-protected
clusters is provided in the Supporting Information.
EXPERIMENTAL SECTION
–
–
The Au25(SC4)18 and Au25(SC2Ph)18 clusters were
prepared and purified as already described.37,42 Full details on
0
chemicals and the preparation of Au24Pt(SC4)18
,
,
0
0
0
Au24Pd(SC4)18
,
Au24Hg(SC4)18
,
Au24Cd(SC4)18
Au24Hg(SC2Ph)180, and Au24Cd(SC2Ph)180 are described in the
Supporting Information.
The UV-vis absorption spectra of the clusters were obtained
in DCM with
a
Thermo Scientific Evolution 60S
spectrophotometer. The spectra were recorded with a spectral
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 14
resolution of 0.5 nm. The samples were at 0.2 mM
concentration in mm cuvettes. MALDI-TOF mass
SHELXT59 and refined by full-matrix least-squares on F2 by
SHELXL60 in the OLEX2 (v. 1.2.10) program.61
1
1
spectrometry experiments were carried out with an Applied
Biosystems 4800 MALDI-TOF/TOF spectrometer equipped
with a Nd:YAG laser operating at 355 nm. The laser-firing
rate was 200 Hz and the accelerating voltage was 25 kV.
trans-2-[3-(4-tert-butylphenyl)-2-methyl-2-propenylidene]
malononitrile (DCTB) was used as the matrix. Depending on
the experiment, the instrument was calibrated with
2
3
4
5
6
7
8
9
ASSOCIATED CONTENT
Supporting Information. Chemicals, syntheses, NMR
tables, further figures (UV-vis absorption spectroscopy
spectra, NMR spectroscopy spectra, MALDI-TOF mass
spectra, DPV experiments), and X-ray crystallographic
analysis of the single-crystal structures of Au24M(SR)18
samples. The CCDC entries 1938192-1938198 contain the
supplementary crystallographic data for this paper. CIFs can
0
0
Au25(SC4)25 or Au25(SC2Ph)250. The clusters were dissolved
in DCM containing DCTB to obtain 0.1 mM solutions with a
1:400 nanocluster/matrix ratio. A 5 l solution was drop cast
onto the sample plate and air-dried. All spectra were recorded
using the reflector positive-ion mode.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
be
obtained
free
of
charge
via
www.ccdc.cam.ac.uk/data_request/cif, or by emailing
data_request@ccdc.cam.ac.uk. The Supporting Information is
The electrochemical experiments were carried out under an
Ar atmosphere, in a glass cell at room temperature, unless
otherwise stated. The solvent-electrolyte system was DCM
containing 0.1 M TBAH. The working electrode was a glassy
carbon microdisk (9.1  10-4 cm2), prepared and activated as
already described.57 As a quasi-reference electrode, we used a
silver wire, which was kept in a tube filled with the same
electrolyte solution and separated from the main compartment
by a Vycor frit. Its calibration was performed by addition of
ferrocene at the end of the experiments; in DCM/0.1 M
TBAH, the ferricenium/ferrocene redox couple has E° = 0.460
V against the KCl saturated calomel electrode (SCE). All
potential values are reported against SCE. The counter-
electrode was a Pt wire. We used a CHI 660c electrochemical
workstation. In CV, we used the positive feedback correction
to minimize the ohmic drop between the working and the
reference electrodes. For DPV, we used peak amplitude of 50
mV, pulse width of 0.05 s, 2 mV increments per cycle, and
pulse period of 0.1 s.
available
on
the
ACS
Publication
website
at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*flavio.maran@unipd.it
ORCID
Sabrina Antonello: 0000-0002-0090-9922
Tiziano Dainese: 0000-0002-5771-7307
Alessandro Dolmella: 0000-0003-1287-6635
Wenwen Fei: 0000-0003-0965-496X
Manu Lahtinen: 0000-0001-5561-3259
Flavio Maran: 0000-0002-8627-6491
Kari Rissanen: 0000-0002-7282-8419
Alfonso Venzo: 0000-0002-3211-0242
Notes
The authors declare no competing financial interest.
1H and 13C NMR spectra were obtained on a Bruker Avance
DMX-600 MHz spectrometer operating at 599.90 and 150.61
MHz, respectively, and equipped with a 5 mm TX-1 inverse
probe powered by field gradients along the x,y,z-axes. The
probe temperature was controlled (0.1 °C) with a Bruker
BVT3000 temperature controller. The chemical shift ()
values are given as ppm downfield from internal
ACKNOWLEDGMENTS
This work was financially supported by the University of
Padova (grant P-DiSC-2017: Gold Nose), Fondazione
CARIPARO (grant: GoldCat), and the University of
Jyväskylä.
1
tetramethylsilane, for both H and 13C nuclei. To ensure a
REFERENCES
1. Antonello, S.; Maran, F. Molecular Electrochemistry of
Monolayer-Protected Clusters. Curr. Opin. Electrochem. 2017, 2,
18−25.
2. Jin, R. Atomically Precise Metal Nanoclusters: Stable Sizes and
Optical Properties. Nanoscale 2015, 7, 1549–1565.
3. Agrachev, M.; Ruzzi, M.; Venzo, A.; Maran, F. Nuclear and
Electron Magnetic Resonance Spectroscopies of Atomically Precise
Gold Nanoclusters. Acc. Chem. Res. 2019, 52, 44−52.
4. Protected Metal Clusters: From Fundamentals to Applications, In
Frontiers of Nanoscience; Tsukuda, T., Häkkinen, H., Eds.; Elsevier:
Amsterdam, 2015; Vol. 9.
5. Jin, R.; Zeng, C.; Zhou, M.; Chen, Y. Atomically Precise
Colloidal Metal Nanoclusters and Nanoparticles: Fundamentals and
Opportunities. Chem. Rev. 2016, 116, 10346-10413.
complete relaxation for all the resonances, the integrals of the
proton spectra were obtained using a pre-scan delay of 10 s.
All measurements were carried out in benzene-d6. The proton
assignments were performed by COSY or TOCSY, whereas
the 13C chemical shift values were obtained from HMQC
experiments. (1H-1H) homodecoupling experiments were
performed with the standard zghd pulse sequence provided in
the Bruker library.
0
Single-crystal X-ray data for the metal doped Au24M(SR)18
clusters were collected either with
a Rigaku Oxford
Diffraction SuperNova dual-source X-ray diffractometer using
hi-flux Mo and Cu micro-focus sources (Mo Kα; λ = 0.71073
Å and Cu Kα; λ = 1.54184 Å) and an Atlas CCD detector
(University of Jyväskylä), and/or with an Oxford Diffraction
Xcalibur Gemini diffractometer with Mo-radiation and Eos
CCD detector (University of Padova). Data collection,
reduction processes, and analytical numeric absorption
corrections by multifaceted crystal models and/or empirical
absorption correction using spherical harmonics, were all
carried out using the program CrysAlisPro (v. 39.46).58
Structures were solved by direct methods with program
6. Jin, R.; Nobusada, K. Doping and alloying in atomically precise
gold nanoparticles. Nano Res. 2014, 7, 285−300.
7. Gan, Z.; Xia, N.; Wu, Z. Discovery, Mechanism, and Application
of Antigalvanic Reaction. Acc. Chem. Res. 2018, 51, 2774–2783.
8. Wang, S.; Li, Q.; Kang, X.; Zhu, M. Customizing the Structure,
Composition, and Properties of Alloy Nanoclusters by Metal
Exchange. Acc. Chem. Res. 2018, 51, 2784–2792.
ACS Paragon Plus Environment

Page 13 of 14
Journal of the American Chemical Society
9. Hossain, S.; Niihori, Y.; Nair, L. V.; Kumar, B.; Kurashige, W.;
Negishi, Y. Alloy Clusters- Precise Synthesis and Mixing Effects.
Acc. Chem. Res. 2018, 51, 3114–3124.
28. Yao, Q.; Yuan, X.; Chen, T.; Leong, D. T.; Xie, J. Engineering
Functional Metal Materials at the Atomic Level. Adv. Mater. 2018,
1802751.
1
2
3
4
5
6
7
8
9
10. Parker, J. F.; Fields-Zinna, C. A.; Murray, R. W. The Story of a
Monodisperse Gold Nanoparticle: Au25L18. Acc. Chem. Res. 2010, 43,
1289−1296.
29. Heaven, M.W.; Dass, A.; White, P.S.; Holt, K.M.; Murray R. W.
Crystal
Structure
of
the
Gold
Nanoparticle
[N(C8H17)4][Au25(SCH2CH2Ph)18]. J. Am. Chem. Soc. 2008, 130,
3754–3755.
11. Kang, X.; Chong, H.; Zhu, M. Au25(SR)18: The Captain of the
Great Nanocluster Ship. Nanoscale 2018, 10, 10758–10834.
12. Qian, H.; Jiang, D.-e.; Li, G.; Gayathri, C.; Das, A.; Gil, R. R.;
Jin, R. Monoplatinum Doping of Gold Nanoclusters and Catalytic
Application. J. Am. Chem. Soc. 2012, 134, 16159-16162.
13. Christensen, S. L.; MacDonald, M. A.; Chatt, A.; Zhang, P.;
Qian, H. F.; Jin, R. C. Dopant Location, Local Structure, and
Electronic Properties of Au24Pt(SR)18 Nanoclusters. J. Phys. Chem. C
2012, 116, 26932−26937.
14. Kwak, K.; Tang, Q.; Kim, M.; Jiang, D.-e.; Lee, D.
Interconversion between Superatomic 6-Electron and 8-Electron
Configurations of M@Au24(SR)18 Clusters (M = Pd, Pt). J. Am.
Chem. Soc. 2015, 137, 10833−10840.
15. Thanthirige, V. F.; Kim, M.; Choi, W.; Kwak, K.; Lee, D.;
Ramakrishna, G. Temperature-Dependent Absorption and Ultrafast
Exciton Relaxation Dynamics in MAu24(SR)18 Clusters (M = Pt, Hg):
Role of the Central Metal Atom. J. Phys. Chem. C 2016, 120,
23180−23188.
16. Tian, S. B.; Liao, L. W.; Yuan, J. Y.; Yao, C. H.; Chen, J. S.;
Yang, J. L.; Wu, Z. K. Structures and Magnetism of Mono-Palladium
and Mono-Platinum Doped Au25(PET)18 Nanoclusters. Chem.
Commun. 2016, 52, 9873−9876.
30. Zhu, M.; Aikens, C. M.; Hollander, F. J.; Schatz, G. C.; Jin, R.
Correlating the Crystal Structure of a Thiol-Protected Au25 Cluster
and Optical Properties. J. Am. Chem. Soc. 2008, 130, 5883-5885.
31. Zhu, M.; Aikens, C. M.; Hendrich, M. P.; Gupta, R.; Qian, H.;
Schatz, G. C.; Jin, R. Reversible Switching of Magnetism in Thiolate-
Protected Au25 Superatoms. J. Am. Chem. Soc. 2009, 131,
2490−2492.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
−
32. Jiang, D.-e.; Dai, S. From Superatomic Au25(SR)18 to
q
Superatomic M@Au24(SR)18 Core-Shell Clusters. Inorg. Chem.
2009, 48, 2720−2722.
33. Walter, M.; Moseler, M. Ligand-Protected Gold Alloy Clusters-
Doping the Superatom. J. Phys. Chem. C 2009, 113, 15834-15837.
34. Taylor, M. G.; Mpourmpakis, G. Rethinking Heterometal Doping
in Ligand-Protected Metal Nanoclusters. J. Phys. Chem. Lett. 2018, 9,
6773-6778.
35. Taylor, M. G.; Mpourmpakis, G. Thermodynamic Stability of
Ligand-Protected Metal Nanoclusters. Nat. Commun. 2017, 8, 15988.
36. Alkan, F.; Pandeya, P.; Aikens, C. M. Understanding the Effect
of Doping on Energetics and Electronic Structure for Au25, Ag25 and
Au38 Clusters. J. Phys. Chem. C 2019, 14, 9516-9527.
37. De Nardi, M.; Antonello, S.; Jiang, D.; Pan, F.; Rissanen, K.;
Ruzzi, M.; Venzo, A.; Zoleo, A.; Maran, F. Gold Nanowired: A
Linear (Au25)n Polymer from Au25 Molecular Clusters. ACS Nano
2014, 8, 8505-8512.
38. Agrachev, M., Antonello, S., Dainese, T., Gascón, J. A., Pan, F.,
Rissanen, K., Ruzzi, M., Venzo, A., Zoleo, A.; Maran F. A Magnetic
Look into the Protecting Layer of Au25 Clusters. Chem. Sci. 2016, 7,
6910-6918.
39. Antonello, S.; Dainese, T.; Pan, F.; Rissanen, K.; Maran, F.
Electrocrystallization of Monolayer Protected Gold Clusters: Opening
the Door to Quality, Quantity and New Structures. J. Am. Chem. Soc.
2017, 139, 4168–4174.
40. Qian, H.; Zhu, M.; Gayathri, C.; Gil, R. R.; Jin, R. Chirality in
Gold Nanoclusters Probed by NMR Spectroscopy. ACS Nano 2011,
11, 8935-8942.
41. Dainese, T.; Antonello, S.; Bogialli, S.; Fei, W.; Venzo, A.;
Maran, F. Gold Fusion: From Au25(SR)18 to Au38(SR)24, the Most
Unexpected Transformation of a Very Stable Nanocluster. ACS Nano
2018, 12, 7057-7066.
42. Dainese, T.; Agrachev, M.; Antonello, S.; Badocco, D.; Black,,
D. M.; Fortunelli, A.; Gascón, J. A.; Stener, M.; Venzo, A.; Whetten,
R. L.; Maran, F. Atomically Precise Au144(SR)60 Nanoclusters (R =
Et, Pr) are Capped by 12 Distinct Ligand Types of 5-fold Equivalence
and Display Gigantic Diastereotopic Effects. Chem. Sci. 2018, 9,
8796-8805.
17. Negishi, Y.; Kurashige, W.; Niihori, Y.; Iwasa, T.; Nobusada, K.
Isolation, Structure, and Stability of a Dodecanethiolate-Protected
Pd1Au24 Cluster. Phys. Chem. Chem. Phys. 2010, 12, 6219−6225.
18. Negishi, Y.; Kurashige, W.; Kobayashi, Y.; Yamazoe, S.;
Kojima, N.; Seto, M.; Tsukuda, T. Formation of a Pd@Au12
Superatomic Core in Au24Pd1(SC12H25)18 Probed by Au197
Mossbauer and Pd K-Edge EXAFS Spectroscopy. J. Phys. Chem.
Lett. 2013, 4, 3579−3583.
19. Tofanelli, M. A.; Ni, T. W.; Phillips, B. D.; Ackerson, C. J.
0
Crystal Structure of the PdAu24(SR)18 Superatom. Inorg. Chem.
2016, 55, 999−1001.
20. Wang, S.; Song, Y.; Jin, S.; Liu, X.; Zhang, J.; Pei, Y.; Meng, X.;
Chen, M.; Li, P.; Zhu, M. Metal Exchange Method Using Au25
Nanoclusters as Templates for Alloy Nanoclusters with Atomic
Precision. J. Am. Chem. Soc. 2015, 137, 4018−4021.
21. Yao, C.; Lin, Y.-j.; Yuan, J.; Liao, L.; Zhu, M.; Weng, L.-h.;
Yang, J.; Wu, Z. Mono-Cadmium vs Mono-Mercury Doping of Au25
Nanoclusters. J. Am. Chem. Soc. 2015, 137, 15350−15353.
22. Zhou, M.; Yao, C.; Sfeir, M. Y.; Higaki, T.; Wu, Z.; Jin, R.
Excited-State Behaviors of M1Au24(SR)18 Nanoclusters: The Number
of Valence Electrons Matters. J. Phys. Chem. Lett. 2018, 122, 13435-
13442.
23. Liao, L.; Zhou, S.; Dai, Y.; Liu, L.; Yao, C.; Fu, C.; Yang, J.;
Wu, Z. Mono-Mercury Doping of Au25 and the HOMO/LUMO
Energies Evaluation Employing Differential Pulse Voltammetry. J.
Am. Chem. Soc. 2015, 137, 9511−9514.
43. Karplus, M. Contact Electron-Spin Coupling of Nuclear
Magnetic Moments. J. Chem. Phys. 1959, 30, 11-15.
44. Zerbe, O.; Pountney, D. L.; von Philipsborn, W.; Vašák, M.
113Cd,1H-Cysteine Coupling in Cd-Substituted Metalloproteins
Follows a Karplus-Type Dependence. J. Am. Chem. Soc. 1994, 116,
377-378.
24. Yan, N.; Liao, L.; Yuan, J.; Lin, Y.-j; Weng, L.-h.; Yang, J.; Wu,
Z. Bimetal Doping in Nanoclusters: Synergistic or Counteractive?
Chem. Mater. 2016, 28, 8240-8247.
25. Yang, S.; Wang, S.; Jin, S.; Chen, S.; Sheng, H.; Zhu, M. A metal
45. Zerbe, O.; Pountney, D. L.; von Philipsborn, W.; Vašák, M.
113Cd,1H-Cysteine Coupling in Cd-Substituted Metalloproteins
Follows a Karplus-Type Dependence. [Erratum to document cited in
CA120:49322] J. Am. Chem. Soc. 1994, 116, 7957-7957.
46. Henehan, C. J.; Pountney, D. L.; Zerbe, O.; Vašák, M.
Identification of Cysteine Ligands in Metalloproteins using Optical
and NMR Spectroscopy: Cadmium-substituted Rubredoxin as a
Model [Cd(CysS)4I2− Center. Protein Sci. 1993, 2, 1756-1764.
47. Pountney, D. L.; Zerbe, O.; von Philipsborn, W.; Egan, J. B.;
0
exchange method for thiolate-protected tri-metal M1AgxAu24−x(SR)18
(M = Cd/Hg) nanoclusters Nanoscale 2015, 7, 10005-10007.
26. Niihori, Y.; Hossain, S.; Sharma, S.; Kumar, B.; Kurashige, W.
Negishi, Y. Understanding and Practical Use of Ligand and Metal
Exchange Reactions in Thiolate-Protected Metal Clusters to
Synthesize Controlled Metal Clusters Exchange reactions in thiolate-
protected metal clusters. Chem. Rec. 2017, 17, 473–484.
27. Zheng, Y.; Jiang, H.; Wang, X. Multiple Strategies for Controlled
Synthesis of Atomically Precise Alloy Nanoclusters. Acta Phys.
Chim. Sin. 2018, 34, 740-754.
3
Vašák, M. J(113Cd,1H) Couplings in Cd(S-Cys) and Cd-2-(S-Cys)-
Cd Moieties Follow a Karplus-like Dependence with the H-C-S-Cd
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 14
Torsion Angle: Application to Protein Structure. Bull. Magn. Reson.
1995, 17, 145–147.
Molecule-like Au38 Nanoparticles. J. Am. Chem. Soc. 2004, 126,
6193-6199.
1
2
3
4
5
6
7
8
48. Donkers, R. L.; Lee. D.; Murray, R. W. Synthesis and Isolation of
the Molecule-like Cluster Au38(PhCH2CH2S)24. Langmuir 2004, 20,
1945-1952.
49. Antonello, S.; Dainese, T.; De Nardi, M.; Perotti, L.; Maran, F.
Insights into the Interface between the Electrolytic Solution and the
Gold Core in Molecular Au25 Clusters. ChemElectroChem. 2016, 3,
1237–1244.
50. Antonello, S.; Holm, A. H.; Instuli, E.; Maran, F. Molecular
Electron-Transfer Properties of Au38 Clusters. J. Am. Chem. Soc.
2007, 129, 9836−9837.
51. Deng, H.; Wang, S.; Jin, S.; Yang, S.; Xu, Y.; Liu, L.; Xiang, J.;
Hu, D.; Zhu. M. Active metal (cadmium) doping enhanced the
stability of inert metal (gold) nanocluster under O2 atmosphere and
the catalysis activity of benzyl alcohol oxidation. Gold Bull. 2015, 48,
161−167.
55. Dainese, T.; Antonello, S.; Gascón, J. A.; Pan, F.; Perera, N. V.;
Ruzzi, M.; Venzo, A.; Zoleo, A.; Rissanen, K.; Maran, F.
Au25(SEt)18,
a Nearly Naked Thiolate-Protected Au25 Cluster:
Structural Analysis by Single Crystal X-ray Crystallography and
Electron Nuclear Double Resonance. ACS Nano 2014, 8, 3904-3912.
56. Du. Y.; Sheng. H.; Astruc. D.; Zhu. M. Atomically Precise Noble
Metal Nanoclusters as Efficient Catalysts: A Bridge between
Structure
and
Properties,
Chem.
Rev.
2019.
DOI:
10.1021/acs.chemrev.8b00726.
9
57. Meneses, A. B.; Antonello, S.; Arevalo, M.-C; Maran, F. Double‐
́
Layer Correction for Electron‐Transfer Kinetics at Glassy Carbon and
Mercury Electrodes in N,N‐Dimethylformamide. Electroanal. 2006,
18, 363−370.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
58. Rigaku Oxford Diffraction, Version 1.171.39.46, 2018.
59. Sheldrick, G. M. SHELXT – Integrated space-group and
crystal-structure determination. Acta Cryst. 2015, A71, 3-8.
60. Sheldrick, G. M. Crystal structure refinement with
SHELXL. Acta Cryst. 2015, C71, 3-8.
61. Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.;
Puschmann, H. OLEX2: a complete structure solution, refinement
and analysis program. J. Appl. Cryst. 2009, 42, 339-341.
52. Antonello, S.; Dainese, T.; Maran, F. Exploring Collective
Substituent Effects: Dependence of the Lifetime of Charged States of
Au25(SCnH2n+1)18 Nanoclusters on the Length of the Thiolate Ligands.
Electroanalysis 2016, 28, 2771-2776.
53. Antonello, S.; Perera, N. V.; Ruzzi, M.; Gascón, J. A.; Maran, F.
Interplay of Charge State, Lability, and Magnetism in the Molecule-
like Au25(SR)18 Cluster. J. Am. Chem. Soc. 2013, 135, 15585–15594
54. Lee, D.; Donkers, R. L.; Wang, G.; Harper, A. S.; Murray, R. W.
Electrochemistry and Optical Absorbance and Luminescence of
For Table of Contents Only
ACS Paragon Plus Environment