Polymorphism of Phosphine-Protected Gold Nanoclusters: Synthesis and Characterization of a New 22-Gold-Atom Cluster

Article abstract of DOI:10.1002/smll.201600407

A new Au₂₂ nanocluster, protected by bis(2-diphenyl-phosphino)ethyl ether (dppee or C₂₈H₂₈OP₂) ligand, has been synthsized and purified with high yield. Electrospray mass spectrometry shows that the new cluster

Full text of DOI:10.1002/smll.201600407

full papers
www.MaterialsViews.com
Gold Nanoclusters
Polymorphism of Phosphine-Protected Gold
Nanoclusters: Synthesis and Characterization of a New
22-Gold-Atom Cluster
Qian-Fan Zhang, Paul G. Williard, and Lai-Sheng Wang*
A new Au22 nanocluster, protected by bis(2-diphenyl-phosphino)ethyl ether (dppee
or C28H28OP2) ligand, has been synthsized and purified with high yield. Electrospray
mass spectrometry shows that the new cluster has a formula of Au22(dppee)7,
containing 22 gold atoms and seven dppee ligands. The cluster is found to be stable
as a solid, but metastable in solution. The new cluster has been characterized by
UV-Vis-NIR absorption spectroscopy, collision-induced dissociation, and 31P-
NMR. The properties of the new cluster have been compared with the previous
Au22(dppo)6 nanocluster (dppo = 1,8-bis(diphenyl-phosphino)octane or C32H36P2),
which contains two fused Au11 units. All the experimental data indicate that the new
Au22(dppee)7 cluster is different from the previously known Au22(dppo)6 cluster and
represents a new Au22 core, which contains most likely one Au11 motif with several
Au2(dppee) or Au(dppee) units. The Au22(dppee)7 cluster provides a new example
of the ligand effects on the nuclearity and structural polymorphism of phosphine-
protected atom-precise gold nanoclusters.
By controlling the size and shape of the core geom-
1. Introduction
etry and the type of ligands, a variety of atomically precise
Ligand-protected gold nanoclusters (AuNCs) have attracted AuNCs may be synthesized with new structures and prop-
tremendous interest recently, due to their potential applica- erties. The type of ligand is especially important, because
tions in catalysis, biosensor, imaging, and molecular elec- of the different metal-ligand interactions. To date, AuNCs
tronics.[1,2] Unlike conventional gold nanoparticles, AuNCs have been synthesized using thiolates/selenolates,[3–23] alky-
consist of only a few or tens of gold atoms protected by nyls,[24–26] phosphines,[27–55] or mixed ligands.[28–31] In par-
ligands. They have atomically precise molecular formulas ticular, a wide range of AuNCs with the thiolate/selenolate
and well-defined coordination geometries. The quantum ligands have been synthesized and structurally character-
[3,4]
[22,23]
confinement effects of such small gold cores lead to discrete ized with gold cores ranging from Au18
to Au133,
electronic states, which give rise to unique molecular-like since the first synthesis and structural determination by
electronic properties for AuNCs. The monodispersity and X-ray crystallography of the Au102 cluster using the para-
well-defined geometry make AuNCs ideal models for under- mercaptobenzoic acid ligand.[20] Thiolate-protected AuNCs
standing structure-property relationships.
have all been found to have complicated gold-ligand inter-
faces, because of the formation of gold-thiolate staples.[20]
From the structural point of view, the ligands that protect
the core are actually gold-thiolate staples, which should be
treated as multidentate ligands. Hence, the true gold cluster
cores are in fact smaller than the AuNC formulas indi-
cate. Similar staples have also been found in the alkynyl-
protected Au19,[24] Au23,[25] and Au24.[26] On the other hand,
phosphine-protected AuNCs tend to have simpler inter-
faces with well-defined cores and ligand layers,[27–31] due
Q. F. Zhang, Prof. P. G. Williard, Prof. L. S. Wang
Department of Chemistry
Brown University
Providence, RI 02912, USA
E-mail: lai-sheng_wang@brown.edu
DOI: 10.1002/smll.201600407
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2518
small 2016, 12, No. 18, 2518–2525
www.small-journal.com

www.MaterialsViews.com
to the neutral charge state of phosphines and the weaker
gold-phosphine bond. The simpler interface and weaker Au-
2. Results
phosphine bond may allow better control of the core size 2.1. Synthesis of Au22(dppee)7
and potentially allow more convenient access to the gold
core in catalytic reactions.
The details of the synthetic and characterization methods
The crystal structures of small phosphine-protected are described in the Experimental Section. The Au22(dppee)7
AuNCs (nuclearity ≤ 13) have been well studied.[27] It is inter- nanoclusters were synthesized in a single step reaction fol-
esting to note that many of these AuNCs, such as Au6,[34–36] lowed by purification, which was different from most thiolate-
Au7,[37,38] Au8,[39,40] Au9,[41,42] Au10,[43,44] and Au11,[45–47] have protected AuNCs and some small size phosphine-protected
at least two different core geometries when different phos- AuNCs.[60,61] The latter all have a special long etching pro-
phine ligands are used. However, beyond Au13, only four dif- cess, in order to obtain the most stable products. However, in
ferent phosphine-protected AuNCs have been structurally the current synthesis of the Au22(dppee)7 nanocluster, longer
characterized by X-ray crystallography, including Au14,[48] etching time did not result in better yield or purity of the
Au20,[49–51] Au22,[52] and Au39.[53] It should be pointed out final product.
that two different Au20 core structures have been reported
using different ligands.[49–51] In addition, two larger AuNCs
with phosphine ligands, i.e., Au55 and Au75,[54,55] have been 2.2. Optimization of Synthetic Conditions Using Electrospray
reported, but they have not been crystallized.
Ionization Mass Spectrometry and UV-Vis-NIR Absorption
The bare Au20 cluster was found in the gas phase to Spectroscopy
have a tetrahedral pyramidal structure (Td).[56] All the 20
Au atoms of the Au20 pyramid are on the cluster surface, We used electrospray ionization mass spectrometry (ESI-
which would make it an ideal model for catalysis. The MS) to monitor and optimize the synthesis. In preparation
pyramidal structure has been observed by high-resolution of the ESI solution, a small amount of formic acid (0.05%)
electron microscopy for size-selected Au20 deposited on was added to facilitate ionization and provided the protons
a carbon substrate in vacuum.[57] The Au20 pyramid was for many ESI-MS peaks in the positive ion mode. Conse-
observed to be present in solution with triphenylphos- quently, the Au22(dppee)7 nanocluster was observed in the
phine (PPh3) ligands using a synthetic procedure for gold ESI-MS in the form of a [Au22(dppee)7H3]3+ ion, similar to
nanoparticles.[58] The pyramidal structure and the electronic our previous observation for the Au22(dppo)6 nanocluster,
properties of the Td Au20(PPh3)4 cluster were shown compu- which was observed in the form of [Au22(dppo)6H4]2+.[52]
tationally to be similar to those of the bare Au20.[58] How- The relative intensity of the [Au22(dppee)7H3]3+ ion peak
ever, in order to obtain large quantities of pure pyramidal in the ESI-MS did not change dramatically, when the reac-
Au20 samples and investigate its interesting catalytic and tion time was increased from 20 hto 120 h, as shown in
optical properties, a high yield synthetic method is needed. Figure 1a, which also revealed mass peaks related to the
One of the strategies that we have been pursuing is to use starting material and a few byproducts. The similar inten-
diphosphine ligands with different chain lengths.[59] Using sity of the [Au22(dppee)7H3]3+ peaks after 20 h and 120 h
the 1,8-bis(diphenyl-phosphino) octane (dppo for short), indicated that the Au22(dppee)7 nanocluster is a relatively
we have synthesized a Au22(dppo)6 cluster with a relatively stable product.
high yield.[52] Structural determination using single crystal
UV-Vis-NIR spectra (Figure 2a) also showed no signifi-
X-ray diffraction revealed that the Au22 core in Au22(dppo)6 cant changes at the 20 h or 120 h reaction times, consistent
contained two fused Au11 units. Interestingly, there are eight with the ESI-MS results. The only differences in the two UV-
Au atoms at the interface of the two Au11 units that were Vis-NIR spectra were the relative intensities of the absorp-
not coordinated by the dppo ligands.
tion peaks at ≈410 and 305 nm, which turned out not to be
Here we report the synthesis of a new Au22 cluster due the absorptions of pure Au22(dppee)7. Figure 2b dis-
using the bis(2-diphenyl-phosphino)ethyl ether (dppee or played the UV-Vis-NIR spectra of pure Au22(dppee)7 dis-
C28H28OP2) ligand, which has a slightly shorter chain length solved in CH2Cl2 after column separation (vide infra). These
than the dppo ligand[52] and can be viewed as replacing spectra were taken for freshly separated samples and at dif-
four CH2 groups in dppo by an ether group. The new Au22 ferent times after the separation, showing that significant
cluster is found to be coordinated by seven dppee ligands, changes occurred after 3 d in solution. The Au22(dppee)7
Au22(dppee)7, and can be obtained with relatively high yield cluster was much more stable as a solid, for which significant
and purity.The yield (~25% based on Au atoms) is higher than changes were only observed after 9 d. The time-dependent
the syntheses of most AuNCs,[3–31] which usually have yields ≤ UV-Vis-NIR spectra suggested that the formation of the
16%. The Au22(dppee)7 cluster has been characterized by UV- Au22(dppee)7 nanocluster might be kinetically controlled. In
Vis-NIR, collision-induced dissociation (CID), and 31P-NMR order to explore the formation kinetics, we carried out syn-
and compared with the previous Au22(dppo)6 cluster.[52] It is theses at different temperatures: −20, 0, 20, and 40 °C. We
concluded that the new Au22(dppee)7 nanocluster has a totally found the best yield was ≈25% at 20 °C, followed by ≈18%
different core structure from the previous Au22(dppo)6 cluster. at 0 and 40 °C, then <7% at −20 °C. The slow reaction rate
The current work provides another example of controlling at low temperatures caused a decrease in yield, while at high
high-nuclearity Au core structures by different ligands and the temperatures the Au22(dppee)7 cluster might be converted to
polymorphism of AuNCs by phosphine ligands.
other products to also give a low yield.
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 2519
small 2016, 12, No. 18, 2518–2525

full papers
www.MaterialsViews.com
+
5
1
1- [(dppee)O H]
2
a)
a)
100
120 hours
1.0
0.8
0.6
0.4
0.2
0.0
+
80
60
40
2- [Au(dppee)]
+
3-
4-
5-
[
[
[
Au(dppee) O
]
2
9
20 hours
120 hours
4
3+
Au (dppee)
13
]
5
3
2
6
20
8
2+
Au (dppee)
]
4
7
8
0
2+
ABS
%
6-
7-
8-
9-
[
[
[
[
Au (dppee) Cl]
11 4
100
80
60
40
20
2+
2+
Au (dppee) O
10
]
20 hours
5
Au (dppee) Cl
13
]
5
3+
Au (dppee) H
]
22
7
3
0
300
400
500
600
700
800
900 1000
0
1000
2000
3000
4000
m/z
wavelength/nm
b)
100
%
b)
9 days in solution
6 days in solution
9 days in solid state
1.2
1.0
0.8
0.6
0.4
0.2
0.0
80
60
40
20
0
under N protection
2
m/z
3 days in solution
fresh sample
0
1000
2000
3000
4000
5000
ABS
simulated [Au(dppee)]+
simulated [Au22(dppee)7H3]3+
100
15
10
5
80
60
40
20
300
400
500
600
700
800
900 1000
638 639 640 641 642 643
2476 2477 2478 2479 2480 2481
wavelength/nm
Figure 1. a) Electrospray mass spectra of the crude products during the
syntheses of Au22(dppee)7 with different reaction times. b) Electrospray
mass spectra of pure Au22(dppee)7 after column separation, as well as
the experimental and simulated isotopic distributions of the two major
peaks. (Mobile phase of ESI: acetonitrle/water = 50/50, 0.05% formic
acid).
Figure 2. a) UV-Vis-NIR of the crude products during the syntheses of
Au22(dppee)7 with different reaction times. b) UV-Vis-NIR spectra of
pure Au22(dppee)7 after column separation in CH2Cl2 solution and the
test of its shelf life. The spectrum of a pure solid sample is also shown
for comparison.
2.3. Separation of the Au22(dppee)7 Nanocluster by
Chromatography
observed after several days, as indicated by the UV-Vis-NIR
spectrum shown in Figure 2b. Due to the stability problem,
we were not able to successfully grow a single crystal of the
The Au22(dppee)7 product could be purified by column chro- Au22(dppee)7 nanocluster, despite of an extensive crystalliza-
matography, as shown in Figure S1 (Supporting Information). tion effort.
The purified sample gave a single mass peak at m/z around
2478 as a triply charged cation in ESI-MS (Figure 1b) and
the simulated isotopic distribution gave the precise formula
of [Au22(dppee)7H3]3+. The UV-Vis-NIR spectrum of the
3. Discussion
freshly purified sample (Figure 2b) displayed distinct absorp- The new Au22(dppee)7 nanocluster has the same gold core
tion peaks at 460, 380, and 310 nm, reflecting the molecular size as the previously reported Au22(dppo)6 cluster.[52] An
nature of the Au22(dppee)7 nanocluster. The distinct and important question was: do they have the same core struc-
well-resolved absorption spectrum also indicated the high ture? Since we were not able to obtain a single crystal for
purity of the column-separated sample. The stability of the the new AuNC, we compared its properties with those of
Au22(dppee)7 nanocluster was tested in solution and the solid the Au22(dppo)6 cluster, in terms of UV-Vis-NIR and 31P
state. The UV-Vis-NIR spectra (Figure 2b) showed the new NMR. To obtain further structural information, we also
AuNC had a finite shelf life in the CH2Cl2 solution with sig- carried out collision-induced dissociation (CID) experi-
nificant changes observed after 3 d. After 6 d in solution, ments and compared the CID characteristics for both
the absorption became significantly broadened and after [Au22(dppee)7H3]3+ and [Au22(dppo)6H4]2+ using tandem
9 d the absorption curve was almost featureless. The puri- ESI-MS. As will be shown below, we found that the Au22
fied Au22(dppee)7 nanocluster was relatively stable in the cores were different in the two AuNCs with the slightly
solid state under nitrogen atmosphere, but change was still different ligands.
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2520
small 2016, 12, No. 18, 2518–2525
www.small-journal.com

www.MaterialsViews.com
3.1. UV-Vis-NIR Spectroscopy of Phosphine-Protected AuNCs
1.0
0.8
0.6
0.4
0.2
0.0
ABS
The UV-Vis-NIR spectra of phosphine-protected AuNCs
are very distinguishable and characteristic of the core size
and structures. The absorption bands between the 300 and
1000 nm UV-Vis-NIR spectral range are related to the elec-
tronic states of the gold core[62] and hence can be used to
derive structural information. The absorption bands from the
phosphine ligands are usually outside the 300−1000 nm spec-
tral range. For gold cores with different nuclearity, or even
the same nuclearity with different structures, the spectra
will show totally different absorption peaks. For example,
although both [Au6(dppp)4]2+ and [Au7(dppp)4]3+ have four
1,3-bis(diphenyl-phosphino)propane (dppp) ligands, their
UV-Vis-NIR spectra show different peaks at 587 and 556 nm,
respectively,[37] due to the different nuclearity. It has also
been reported that the [Au6(dppp)4]2+, [Au8(dppp)4Cl2]2+,
and [Au11(dppp)6]3+ clusters all have structures consisting of
a gold core with addition atoms outside the core (core+exo)
Au22(dppee)7
Au22(dppo)6
300
400
500
600
700
800
900 1000
Wavelength/nm
Figure 3. Comparison of the UV-Vis-NIR spectra of Au22(dppee)7 and
Au22(dppo)6.
and they all have different absorption features, when com- test this hypothesis, we compared the UV-Vis-NIR spectra
pared with the core-only Au nanoclusters of the same of the isolated Au11 nanocluster,[46,64] the previously reported
nuclearity.[45,63] On the other hand, for the same gold core [Au20(PPhPy2)10Cl4]Cl2 nanocluster that contains two Au11
structure, or even different gold cores that share the same units sharing two Au atoms,[49] and the two Au22 nanoclus-
gold building units, the UV-Vis spectra always show similar ters in Table 1. The Au11 clusters protected by PPh3 or PPh3/
absorption features, regardless of the phosphine ligands. Cl display strong absorption bands at ≈420 nm and ≈310 nm,
For example, the Au11(PPh3)7Cl3 and [Au11(PPh3)8Cl2]Cl in addition to weak absorptions at 515 nm and ≈380 nm. The
clusters have the same Au11 core, but have different num- two strong absorption bands are seen to be preserved in the
bers of Cl and PPh3 ligands.[46,64] Their UV-Vis-NIR spectra Au22(dppo)6 and [Au20(PPhPy2)10Cl4]Cl2 AuNCs, although
are very similar and show strong absorption peaks at ≈310 slightly shifted because the latter two clusters contain fused
and 420 nm, and weak peaks at ≈380 and 515 nm, as docu- Au11 units. These two absorption bands are also observed in
mented in Table 1. Another example is the recently reported the new Au22(dppee)7 cluster, suggesting it should also con-
[Au20(PP3)4]Cl4 (PP3
= tris[2-(diphenyl-phosphino)ethyl] tain an Au11 unit. In addition, its strong absorption at 380 nm
phosphine, C42H42P4) cluster,[50,51] which consists of an ico- is also similar to that in the two Au11 clusters. The large shift
sahedral Au13 building unit with an extra Au7 clamp as a of the 460 nm band in Au22(dppee)7 relative to the ≈420 nm
new layer. It was observed that the [Au20(PP3)4]Cl4 cluster band in the Au11 cluster suggests that the new Au22 cluster
gave similar UV-Vis-NIR spectral features as the previously core is unlikely to be a polymer[66] or assembly[67] of isolated
known icosahedral cluster, [Au13(dppe)5Cl2]Cl3 (dppe = dppee-protected Au11 units. The CID and NMR data to be
1,2-bis(diphenyl-phosphino)ethane, C26H24P2) at ≈360 and presented below are consistent with these observations.
≈490 nm.[65]
The UV-Vis-NIR spectrum of Au22(dppee)7 is com-
pared with that of Au22(dppo)6 in Figure 3 and they are not 3.2. Collision-Induced Dissociations
identical, suggesting the core geometry in the two AuNCs
cannot be the same. However, the two spectra share a sim- To obtain further structural information, we carried out
ilar strong absorption band at ≈460 nm, as well as a band at CID experiments of Au22(dppee)7 and compared them
≈310 nm. Since the Au22(dppo)6 cluster consists of two fused with those of Au22(dppo)6. In the positive-ion ESI-MS,
Au11 units, the similar absorption features suggest the new many mass peaks were observed as protonated spe-
Au22(dppee)7 nanocluster may also contain an Au11 unit. To cies because trace amount of formic acid was added to
the mobile phase to facilitate the ESI (see the Experi-
mental Section). Specifically, the Au22(dppo)6 cluster was
Table 1. Comparison of the main absorption bands in the UV-Vis-NIR
spectra of Au22(dppee)7, Au11(PPh3)7Cl3, [Au11(PPh3)8Cl2]Cl,
Au22(dppo)6, and [Au20(PPhPy2)10Cl4]Cl2. Strong peaks are in bold.
observed previously in the form of [Au22(dppo)6H4]2+ (as
well as alkali-complexed cations),[52] whereas the cur-
rent Au22(dppee)7 cluster was observed in the form of
[Au22(dppee)6H3]3+ (Figure 1). The proton was most likely
associated with the P atom on the dppo ligands or the P
and/or O atoms in the case of the dppee ligands. As shown
in Figure 4, at low collision energies the CID of both Au22
nanoclusters showed very strong intensity (nearly 100%)
for the parent molecular ions with relatively minor disso-
ciations, indicating the thermal dynamic stability of both
Wavelength of the peaks [nm]
Reference
Current work
[46,64]
[46,64]
[52]
Au22(dppee)7
–
515
515
–
460
420
416
456
493
380
385
380
–
310
308
312
313
344
Au11(PPh3)7Cl3
[Au11(PPh3)8Cl2]Cl
Au22(dppo)6
[Au20(PPhPy2)10Cl4]Cl2
–
–
[49]
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 2521
small 2016, 12, No. 18, 2518–2525

full papers
www.MaterialsViews.com
100
80
60
40
20
%
Au22(dppee)7 cluster was more stable
than Au22(dppo)6.
a)
The major dissociation channels of
Au22(dppee)7 involved the loss of dppee,
Au(dppee), Au(dppee)2, and Au2(dppee)2.
At a given collision energy, the parent
ion of Au22(dppo)6 displayed much
less degree of dissociation (Figure 4b;
Figure S3, Supporting Information). The
different CID behaviors of the parent
ions of Au22(dppee)7 and Au22(dppo)6
also suggested that they have different
core structures. It should be pointed out
that the CID of the phosphine ligand-
protected Au11 cluster showed completely
different behaviors from the two Au22
clusters.[68,69] The CID experiments are
consistent with the UV-Vis-NIR data that
the Au22(dppee)7 cluster may contain an
Au11 unit, but with additional Au atoms
coordinated around it. The 31P-NMR data
below provide similar structural hints for
Au22(dppee)7.
[Au22(dppee)6H]3+
[Au22(dppee)7H3]3+
[Au(dppee)]+
35V
25V
20V
15V
10V
0
1000
2000
3000
4000
m/z
100
80
60
40
20
0
%
b)
[Au22(dppo)6H4]2+
[Au22(dppo)5]2+
[Au(dppo)]+
3.3. 31P-NMR
We have also used the 31P NMR to char-
acterize the new Au22(dppee)7 cluster.
Although 31P-NMR cannot give definitive
structures, it is characteristic for different
phosphine-protected AuNCs. For example,
two icosahedral Au13 nanoclusters with
different phosphine ligands[65,70] both
show a single sharp 31P peak at 66–67 ppm.
The Au20 cluster with the Au13 building
unit[50,51] also has a strong 31P peak at
66–67 ppm. Another nanocluster with the
Au13 building unit was Au19, protected by
35V
25V
15V
0
1000
2000
3000
4000
m/z
Figure 4. Collision-induced dissociation of a) [Au22(dppee)7H3]3+ and b) [Au22(dppo)6H4]2+ at
different collision energies. (Mobile phase of ESI: acetonitrle/water = 50/50, 0.05% formic
acid).
N,N-bis(diphenylphosphino)amine
and
clusters. The major fragmentation channel in both cases alkynylbenzene.[24] The single 31P peak in Au19 was observed
was the loss of a gold atom and a ligand (L), i.e., AuL+. at 76 ppm; the shift was due to the stronger electron-with-
It should be pointed out that the remaining heavier frag- drawing effect of the nitrogen atom. As for the nanoclus-
ment after the loss of an AuL+ unit would be beyond the ters with the Au11 building unit, such as Au11(PPh3)7Cl3,
mass range of our instrument and could not be detected [Au11(PPh3)8Cl2]Cl, and [Au20(PPhPy2)10Cl4]Cl2, they all
in the tandem MS-MS mode. When the collision energy have a single strong 31P peak at 52–53 ppm.[46,49]
was increased, the mass spectra of both Au22 nanoclusters
Figure 5 shows the 31P-NMR spectrum of Au22(dppo)6
show weaker parent ion peaks, along with more fragmen- in comparison with that of Au22(dppee)7. In both spectra,
tation. The relative intensities of the different fragments as numerous broad peaks were observed at 44–54 ppm, which
a function of collision energies are given in Figures S2 and were in the same range of chemical shifts of the phosphine
S3 (Supporting Information) for [Au22(dppee)6H3]3+ and ligands on the Au11 building unit. The multiplicity and broad-
[Au22(dppo)6H4]2+, respectively. The [Au22(dppee)6H3]3+ ening of the peaks could be explained by the relatively
parent ion was observed to be much less stable than the slow exchange rate between different 31P nuclei. In prin-
[Au22(dppo)6H4]2+ ion and was almost completely dissoci- ciple, all the phosphorous atoms coordinated to the Au11
ated above the collision energy of 20 V (Figure S2, Sup- unit should have different chemical shifts, because of the
porting Information). In addition, many more fragments asymmetric gold core. However, the fast exchange of small
were observed for the [Au22(dppee)6H3]3+ parent ion than ligands like PPh3 and PPhPy2 makes these 31P nuclei indis-
for the [Au22(dppo)6H4]2+ ion. Hence, in the gas-phase tinguishable. Thus, only a single sharp peak was observed
protonated forms, the CID experiments indicated that the previously for Au11(PPh3)7Cl3, [Au11(PPh3)8Cl2]Cl, and
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2522
small 2016, 12, No. 18, 2518–2525
www.small-journal.com

www.MaterialsViews.com
Supporting Information). Since the only byproduct of dppee
that gave similar chemical shifts after the column separa-
tion was dppee oxides, the 28 ppm peak in the 31P-NMR of
Au22(dppee)7 was also likely due to dppee oxides, similar to
the ligand oxide in the Au22(dppo)6 cluster.
The peak at 25 ppm was a new feature that was not
observed in the 31P-NMR of Au22(dppo)6. It could be
assigned to Au2Cl2(dppee) by comparing with the 31P-NMR
of Au2Cl2(dppee) (Figure S10, Supporting Information).
However, unlike the sharp 28 ppm peak, the 25 ppm peak
in 31P-NMR of Au22(dppee)7 appeared as a doublet, which
became sharper and more intense at low temperatures. The
temperature effect of the 25 ppm doublet was similar to
the peaks at 44–54 ppm, suggesting that it should belong to
the Au22(dppee)7 nanocluster itself. These observations sug-
gested that the Au22(dppee)7 cluster may contain motifs like
Au2(dppee) at different positions. This conclusion is con-
Figure 5. Comparison of the 31P-NMR of Au22(dppo)6 and Au22(dppee)7
at 293 K.
[Au20(PPhPy2)10Cl4]Cl2.[46,49] Since the migration of bulky sistent with the structural inferences from both the UV-Vis-
long-chain ligands like dppee and dppo were much slower, NIR and the CID results, that is, the Au22(dppee)7 cluster
multiple broad peaks were observed in Au22(dppo)6 and should contain an Au11 unit with additional coordination of
Au22(dppee)7. Additionally, the coupling between these Au2(dppee)/Au(dppee), which are likely in the exterior of
chemically different 31P nuclei could also contribute to the the Au11 unit.
multiplicity and broadening. In order to further explore
the existence of the dynamic behavior of the 31P nuclei in
Au22(dppee)7, we measured its 31P-NMR at different temper- 3.4. Summary of Structural Features of the Au22(dppee)7
atures, as shown in Figure 6 (the chemical shifts of all peaks Nanocluster
are given in Figures S4–S6, Supporting Information). At low
temperatures, the exchange rate was slowed down, so that the The UV-Vis-NIR, CID, and 31P-NMR results are all consistent
31P-NMR peaks became sharper and better resolved. On the with the existence of an Au11 unit and additional Au2(dppee)/
other hand, at high temperatures one expected the chemical Au(dppee) in the Au22(dppee)7 nanocluster. However, unlike
shifts to coalesce, even though we did not try temperatures thiolate or alkynyl-AuNCs, there are unlikely Au-dppee sta-
higher than 298 K due to concerns of decomposition.
ples in Au22(dppee)7. In the thiolate- or alkynyl-AuNCs, the
In comparison, the 31P-NMR of Au22(dppo)6 (Figure 5) ligands donate two pairs of electrons to Au,[72] one to the gold
also showed a single peak at ≈31 ppm, which was most likely core and one to the Au atom in the staple. However, the P
due to the dppo oxide.[71] In the above CID study, we showed atom in phosphine ligands only has one pair of electrons
that the loss of a dppo ligand was the first dissociation available, and can only coordinate to one Au atom. Hence,
channel for Au22(dppo)6 in the gas phase (Figure 4b). It was the additional Au atoms in Au22(dppee)7 other than the Au11
possible that in solution Au22(dppo)6 could also lose a ligand, unit should all form Au-Au bonds and are likely bonded to
which was then quickly oxidized to give the 31 ppm peak. In each other in some form, either as a new layer or as smaller
the 31P-NMR of Au22(dppee)7, there were additional peaks at clusters linked to the Au11 unit. On the basis of the stability
≈28 and 25 ppm.We observed that the same 28 ppm peak also of the Au22(dppee)7 cluster, we suspected that it might also
appeared in the 31P-NMR of dppee as an impurity (Figure S7, contain uncoordinated Au atoms on the core surface, like
those in Au22(dppo)6.
4. Conclusion
A new Au22(dppee)7 nanocluster has been synthesized with
high purity and high yield. The dppee ligand is an analogue
of the previous long-chain dppo ligand.52 Characteristic
UV-Vis-NIR absorption peaks at 460, 380, and 310 nm were
observed for the Au22(dppee)7 cluster. Its precise molecular
formula was determined by high-resolution electrospray
mass spectrometry. The new AuNC was observed to have a
shelf life of three to six days, making it difficult to grow single
crystals. By comparing the UV-Vis-NIR, collision-induced
dissociations, 31P-NMR spectra of the new Au22(dppee)7,
Figure 6. The 31P-NMR of Au22(dppee)7 at 298, 253, and 193 K. Detailed
chemical shifts are given in Figures S4–S6 (Supporting Information).
and the previous Au22(dppo)6 cluster, we concluded that
Au22(dppee)7 has a different core structure, consisting of an
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 2523
small 2016, 12, No. 18, 2518–2525

full papers
www.MaterialsViews.com
Au11 unit with additional Au atoms most likely as a new layer vacuum for 12 h and stored under the protection of nitrogen.
on its surface. The current Au22(dppee)7 cluster and the pre- NMR was used to characterize the product (Figures S10−S14, Sup-
vious Au22(dppo)6 cluster reveal that the Au22 nanoclusters porting Information).
exhibit interesting core isomers, depending on the ligands.
Synthesis of Au22(dppee)7 Nanocluster: Au2Cl2(dppee)
Such structural polymorphism of phosphine-coordinated (102 mg) was dissolved in 85 mL DCM. After vigorous stirring for
atom-precise gold clusters is intriguing and provides excel- 10 min at 20 °C, 3.5 mL ethanol solution of sodium borohydride
lent systems for structure-property correlations.
(made by dissolving 41 mg sodium borohydride in 10 mL ethanol)
was quickly added to the system. The reaction was allowed to pro-
ceed for 20 h before the removal of all solvents. Then the product
was extracted twice by DCM (10 mL). The DCM extractions were
combined, filtered and evaporated to dryness. The crude product
was then purified by silica gel column chromatography (DCM/ethyl
acetate/methanol = 8/2/3) to obtain a pure Au22(dppee)7 sample
with a yield of ≈25% based on Au atoms.
5. Experimental Section
Chemicals: All chemicals were obtained from Sigma-Aldrich,
and were used as received unless further noted. All solvents were
technical grade unless further stated. All reactions under anhydrous
conditions were conducted by using flame- or oven-dried glassware
and standard syringe techniques under nitrogen atmosphere.
Characterizations: NMR spectra were recorded on both a 400
and a 600 MHz Bruker Ultrashield spectrometer using CDCl3 or
CD2Cl2 as solvent. Chemical shifts are reported in ppm and are
referenced to tetramethylsilane (internal) for 1H or 13C-NMR and
85% H3PO4 (external) for 31P-NMR. Unless further noted, all NMR
spectra were measured at 298 K. UV–Vis-NIR spectra were meas-
ured in CH2Cl2 solution on a Varian Cary 50 Bio spectrophotometer.
The mass spectra were measured in the positive ionization mode.
The CH2Cl2 solution of AuNCs (≈0.5 mg mL−1) was introduced into
an Agilent 6530 Accurate Mass Q-TOF LC-MS system (with Agilent
1260 HPLC) via flow injection. The ESI mobile phase was made
of acetonitrile/water (50/50) with ≈0.05% formic acid at a flow
rate of 200 µL min−1. The gas temperature of the ESI source was
130 °C at a flow rate of 8 L min−1. The fragmentor voltage was set
at 50 V, skimmer at 65 V, and the capillary voltage (Vcap) at 3500 V.
The ESI-MS and collision-induced dissociation (CID) spectra were
recorded at a rate of 1 spectrum s−1. For the ESI-MS-CID experi-
ment, the precursor parent ion was isolated with an m/z width of
≈4. The ESI-MS-CID spectra were acquired at collision energies of
15, 25, and 35 V. The mass range measured was up to 20 000 for
MS and 4000 for MS/MS. The assignments were based on high-
resolution m/z values and isotopic distributions.
Supporting Information
Supporting Information is available from the Wiley Online Library
or from the author.
Acknowledgements
The authors would like to thank Dr. Tun-Li Shen for his help with the
ESI-MS-CID experiments and Onkei Tai for her assistance with the
low temperature NMR experiments.
[1] M. T. Pradeep, Part. Part. Syst. Char. 2014, 31, 1017.
[2] T. Tsukuda, H. Häkkinen, Protected Metal Clusters: From Funda-
mentals to Applications, Elsevier Science, Waltham, MA, USA,
2015, 1–385.
[3] A. Das, C. Liu, H. Y. Byun, K. Nobusada, S. Zhao, N. Rosi, R. C. Jin,
Angew. Chem. Int. Ed. 2015, 54, 3140.
Synthesis of the [bis(2-Diphenylphosphino)ethyl]Ether (dppee,
C28H28OP2) Ligand: 2-chloro-ethyl ether (2.5 mmol) was added
to a 100 mL dry Schlenk flask. 20 mL tetrahydrofuran (THF) from
a dry solvent dispensing system was then added to the Schlenk
flask. After 15 min of vigorous stirring, potassium diphenylphos-
phide solution (10.5 mL, 0.5 M in dry THF) was slowly added to the
system. The reaction was allowed to proceed for 24 h at 30 °C. After
the removal of all solvents by a rotary evaporator, the product was
extracted twice by 10 mL dichloromethane (DCM). The DCM extrac-
tions were combined and filtered. The filtrate was evaporated to
give crude product. The pure product (≈75% yield) was obtained
by silica gel column chromatography (hexanes/DCM), and charac-
terized by NMR (Figures S7−S9, Supporting Information).
Synthesis of bis[chlorogold(I)][dppee] (Au2Cl2(dppee), C28H28O-
P2Au2Cl2) : dppee (1 mmol) was added to a dry 50 mL Schlenk flask.
DCM (5 mL) from a dry solvent dispensing system was added to
the flask. After 20 min of vigorous stirring, chloro(dimethylsulfide)-
gold(I) solution (20 mL, 0.1 mmol mL−1 in dry DCM) was quickly
added to the system. The reaction was allowed to proceed for 36 h
at 20 °C. After the removal of all solvents by a rotary evaporator,
the product (≈95% yield based on Au atoms) was dried under
[4] S. Chen, S. X. Wang, J. Zhong, Y. B. Song, J. Zhang, H. T. Sheng,
Y. Pei, M. Z. Zhu, Angew. Chem. Int. Ed. 2015, 54, 3145.
[5] J. Zeng, C. Liu, Y. X. Chen, N. L. Rosi, R. C. Jin, J. Am. Chem. Soc.
2014, 136, 11922.
[6] A. Das, T. Li, K. Nobusada, C. J. Zeng, N. L. Rosi, R. C. Jin, J. Am.
Chem. Soc. 2013, 135, 18264.
[7] A. Das, T. Li, G. Li, K. Nobusada, C. J. Zeng, N. L. Rosi, R. C. Jin,
Nanoscale 2014, 6, 6458.
[8] D. Crasto, G. Barcaro, M. Stener, L. Sementa, A. Fortunelli,
A . D a s s , J. Am. Chem. Soc. 2014, 136, 14933.
[9] Y. B. Song, S. X. Wang, J. Zhang, X. Kang, S. Chen, P. Li,
H. T. Sheng, M. Z. Zhu, J. Am. Chem. Soc. 2014, 136, 2963.
[10] M. W. Heaven, A. Dass, P. S. White, K. M. Holt, R. W. Murray, J. Am.
Chem. Soc. 2008, 130, 3754.
[11] M. Zhu, C. M. Aikens, F. J. Hollander, G. C. Schatz, R. Jin, J. Am.
Chem. Soc. 2008, 130, 5883.
[12] M. Z. Zhu, W. T. Eckenhoff, T. Pintauer, R. C. Jin, J. Phys. Chem.
C 2008, 112, 14221.
[13] J. Zeng, T. Li, A. Das, N. L. Rosi, R. C. Jin, J. Am. Chem. Soc. 2013,
135, 10011.
[14] H. Y. Yang, Y. Wang, A. J. Edwards, J. Z. Yan, N. F. Zheng, Chem.
Commun. 2014, 50, 14325.
[15] D. Crasto, S. Malola, G. Brosofsky, A. Dass, H. Hakkinen, J. Am.
Chem. Soc. 2014, 136, 5000.
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2524
small 2016, 12, No. 18, 2518–2525
www.small-journal.com

www.MaterialsViews.com
[16] A. Das, C. Liu, C. J. Zeng, G. Li, T. Li, N. L. Rosi, R. C. Jin, J. Phys. [42] J. M. Smits, P. Beurskens, J. Bour, F. Vollenbroek, J. Crystallogr.
Chem. A 2014, 118, 8264.
[17] S. Yang, J. S. Chai, Y. B. Song, X. Kang, H. T. Sheng, H. B. Chong, [43] J. Pethe, C. Maichle-Mössmer, J. Strähle, Z. Anorg. Allg. Chem.
M. Z. Zhu, J. Am. Chem. Soc. 2015, 137, 10033. 1998, 624, 1207.
[18] H. F. Qian, W. T. Eckenhoff, Y. Zhu, T. Pintauer, R. C. Jin, J. Am. [44] M. T. Cheetham, M. M. Harding, J. L. Haggitt, D. M. P. Mingos,
Chem. Soc. 2010, 132, 8280. H. R. Powell, J. Chem. Soc., Chem. Commun. 1993, 1000.
[19] M. Azubel, J. Koivisto, S. Malola, D. Bushnell, G. L. Hura, [45] Y. Shichibu, Y. Kamei, K. Konishi, Chem. Commun. 2012, 48,
Spectrosc. Res. 1983, 13, 365.
A. L. Koh, H. Tsunoyama, T. Tsukuda, M. Pettersson, H. Hakkinen,
R. D. Kornberg, Science 2014, 345, 909.
[20] P. D. Jadzinsky, G. Calero, C. J. Ackerson, D. A. Bushnell,
R. D. Kornberg, Science 2007, 318, 430.
[21] Y. X. Chen, C. J. Zeng, C. Liu, K. Kirschbaum, C. Gayathri, R. R. Gil,
N. L. Rosi, R. C. Jin, J. Am. Chem. Soc. 2015, 137, 10076.
[22] A. Dass, S. Theivendran, P. R. Nimmala, C. Kumara, V. R. Jupally,
7559.
[46] L. C. McKenzie, T. O. Zaikova, J. E. Hutchison, J. Am. Chem. Soc.
2014, 136, 13426.
[47] R. C. B. Copley, D. M. P. Mingos, J. Chem. Soc., Dalton Trans. 1996,
479.
[48] S. Gutrath, I. M. Oppel, O. Presly, I. Beljakov, V. Meded, W. Wenzel,
U. Simon, Angew. Chem. Int. Ed. 2013, 52, 3529.
A. Fortunelli, L. Sementa, G. Barcaro, X. B. Zuo, B. C. Noll, J. Am. [49] X. K. Wan, Z. W. Lin, Q. M. Wang, J. Am. Chem. Soc. 2012, 134,
Chem. Soc. 2015, 137, 4610. 14750.
[23] A. Zeng, Y. Chen, K. Kirschbaum, K. Appavoo, M. Y. Sfeir, R. Jin, [50] J. Chen, Q. F. Zhang, P. G. Williard, L. S. Wang, Inorg. Chem. 2014,
Sci. Adv. 2015, 1, e1500045. 53, 3932.
[24] X. K. Wan, Q. Tang, S. F. Yuan, D. E. Jiang, Q. M. Wang, J. Am. [51] X. K. Wan, S. F. Yuan, Z. W. Lin, Q. M. Wang, Angew. Chem. Int. Ed.
Chem. Soc. 2015, 137, 652. 2014, 53, 2923.
[25] X. K. Wan, S. F. Yuan, Q. Tang, D. E. Jiang, Q. M. Wang, Angew. [52] J. Chen, Q. F. Zhang, T. A. Bonaccorso, P. G. Williard, L. S. Wang,
Chem. Int. Ed. 2015, 54, 5977. J. Am. Chem. Soc. 2014, 136, 92.
[26] X. K. Wan, W. W. Xu, S. F. Yuan, Y. Gao, X. C. Zeng, Q. M. Wang, [53] B. K. Teo, X. Shi, H. Zhang, J. Am. Chem. Soc. 1992, 114, 2743.
Angew. Chem. Int. Ed. 2015, 54, 9683.
[54] G. Schmid, R. Pfeil, R. Boese, F. Bandermann, S. Meyer, G. Calis,
[27] K. Konishi, in Gold Clusters, Colloids and Nanoparticles I, Vol. 161
J. W. A. van der Velden, Chem. Ber. 1981, 114, 3634.
(Ed.: D. M. P. Mingos, Berlin, Germany), Springer International [55] E. Gutiérrez, R. D. Powell, F. R. Furuya, J. F. Hainfeld, T. G. Schaaff,
Publishing, 2014, p. 49–86.
[28] A. Das, T. Li, K. Nobusada, Q. Zeng, N. L. Rosi, R. C. Jin, J. Am.
Chem. Soc. 2012, 134, 20286.
M. N. Shafigullin, P. W. Stephens, R. L. Whetten, Eur. Phys. J. D
1999, 9, 647.
[56] J. Li, X. Li, H.-J. Zhai, L.-S. Wang, Science 2003, 299, 864.
[29] H. F. Qian, W. T. Eckenhoff, M. E. Bier, T. Pintauer, R. C. Jin, Inorg. [57] Z. W. Wang, R. E. Palmer, Nanoscale 2012, 4, 4947.
Chem. 2011, 50, 10735.
[58] H.-F. Zhang, M. Stender, R. Zhang, C. Wang, J. Li, L.-S. Wang,
J. Phys. Chem. B 2004, 108, 12259.
[59] M. F. Bertino, Z.-M. Sun, R. Zhang, L.-S. Wang, J. Phys. Chem.
B 2006, 110, 21416.
[30] R. X. Jin, C. Liu, S. Zhao, A. Das, H. Z. Xing, C. Gayathri, Y. Xing,
N. L. Rosi, R. R. Gil, R. C. Jin, ACS Nano 2015, 9, 8530.
[31] Y. B. Song, F. Y. Fu, J. Zhang, J. S. Chai, X. Kang, P. Li,
S. L. Li, H. P. Zhou, M. Z. Zhu, Angew. Chem. Int. Ed. 2015, 54, [60] Z. T. Luo, V. Nachammai, B. Zhang, N. Yan, D. T. Leong, D. E. Jiang,
8430. J. P. Xie, J. Am. Chem. Soc. 2014, 136, 10577.
[32] Z. L. Wu, D. E. Jiang, A. K. P. Mann, D. R. Mullins, Z. A. Qiao, [61] M. Pettibone, J. W. Hudgens, ACS Nano 2011, 5, 2989.
L. F. Allard, C. J. Zeng, R. C. Jin, S. H. Overbury, J. Am. Chem. Soc. [62] R. C. Jin, Nanoscale 2015, 7, 1549.
2014, 136, 6111.
[33] G. Li, H. Abroshan, Y. Chen, R. Jin, H. J. Kim, J. Am. Chem. Soc. [64] X. Kang, Y. B. Song, H. J. Deng, J. Zhang, B. J. Liu, C. S. Pan,
2015, 137, 14295. M. Z. Zhu, RSC Adv. 2015, 5, 66879.
[34] E. S. Smirnova, A. M. Echavarren, Angew. Chem. Int. Ed. 2013, 52, [65] Y. Shichibu, K. Konishi, Small 2010, 6, 1216.
[63] Y. Shichibu, K. Konishi, Inorg. Chem. 2013, 52, 6570.
9023.
[66] M. De Nardi, S. Antonello, D. E. Jiang, F. F. Pan, K. Rissanen,
[35] E. Briant, K. P. Hall, D. M. P. Mingos, A. C. Wheeler, J. Chem. Soc.,
Dalton Trans. 1986, 687.
[36] J. W. A. van der Velden, J. J. Bour, J. J. Steggerda, P. T. Beurskens,
M. Roseboom, J. H. Noordik, Inorg. Chem. 1982, 21, 4321.
M . R u z z i , A . V e n z o , A . Z o l e o , F . M a r a n , ACS Nano 2014, 8, 8505.
[67] W. S. Compel, O. A. Wong, X. Chen, C. Yi, R. Geiss, H. Häkkinen,
K. L. Knappenberger, C. J. Ackerson, ACS Nano 2015, 9, 11690.
[68] E. Bergeron, J. W. Hudgens, J. Phys. Chem. C 2007, 111, 8195.
[37] Y. Shichibu, M. Z. Zhang, Y. Kamei, K. Konishi, J. Am. Chem. Soc. [69] P. S. D. Robinson, T. L. Nguyen, H. Lioe, R. A. J. O’Hair,
2014, 136, 12892. G. N. Khairallah, Int. J. Mass Spectrom. 2012, 330, 109.
[38] J. W. A. van der Velden, P. T. Beurskens, J. J. Bour, W. P. Bosman, [70] M. Sugiuchi, Y. Shichibu, T. Nakanishi, Y. Hasegawa, K. Konishi,
J. H. Noordik, M. Kolenbrander, J. A. K. M. Buskes, Inorg. Chem.
1984, 23, 146.
[39] Y. Kamei, Y. Shichibu, K. Konishi, Angew. Chem. Int. Ed. 2011, 50,
7442.
Chem. Commun. 2015, 51, 13519.
[71] P. Calcagno, B. M. Kariuki, S. J. Kitchin, J. M. A. Robinson, D. Philp,
K. D. M. Harris, Chem. Eur. J. 2000, 6, 2338.
[72] D. M. P. Mingos, Dalton Trans. 2015, 44, 6680.
[40] J. W. A. van der Velden, J. J. Bour, W. P. Bosman, J. H. Noordik,
J. Chem. Soc., Chem. Commun. 1981, 1218.
[41] K. P. Hall, B. R. C. Theoblad, D. I. Gilmour, D. M. P. Mingos,
A. J. Welch, J. Chem. Soc., Chem. Commun. 1982, 528.
Received: February 5, 2016
Revised: February 23, 2016
Published online: March 23, 2016
© 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.small-journal.com 2525
small 2016, 12, No. 18, 2518–2525