Oxothiomolybdenum derivatives of the superlacunary crown heteropolyanion {P8W48}: Structure of [K4{Mo4O 4S4(H2O)3(OH)2} 2-(WO2)(P8W48O184)] 30- and studies in solution

Article abstract of DOI:10.1021/ic202346r

Reaction of the cyclic lacunary [H₇P₈W ₄₈O₁₈₄]^30- anion (noted P₈W ₄₈)with the [Mo2S2O2(H₂O)₆]²⁺ oxothiocation led to two compounds, namely, [K₄{Mo₄O ₄S₄(H₂O)₃-(OH)₂} ₂(WO₂)(P₈W₄₈O₁₈₄)] ^30- (denoted 1)and [{Mo₄O₄S₄(H ₂O)₃(OH)₂}₂(P₈W ₄₈O₁₈₄)]36-(denoted 2), which were characterized in the solid state and solution. In the solid state, the structure of [K ₄{Mo₄O₄S₄(H₂O) ₃(OH)₂}₂-(WO₂)(P₈W ₄₈O₁₈₄)]^30- reveals the presence of two disordered {Mo₄O₄S₄(H₂O) ₃(OH)₂}₂+ "handles" connected on both sides of the P₈W₄₈ ring. Such a disorder is consistent with the presence of two geometrical isomers where the relative disposition of the two {Mo₄O₄S₄(H₂O) ₃(OH)₂}₂+ handles are arranged in a perpendicular or parallel mode. Such an interpretation is fully supported by 31P and 183W NMR solution studies. The relative stability of both geometrical isomers appears to be dependent upon the nature of the internal alkali cations, i.e., Na+ vs K+, and increased lability of the two {Mo₄O ₄S₄(H₂O)₃(OH)₂} ₂+ handles, compared to the oxo analogous, was clearly identified by significant broadening of the 31P and 183W NMR lines. Solution studies carried out by UV-vis spectroscopy showed that formation of the adduct [{Mo ₄O₄S₄(H₂O)₃(OH) ₂}₂(P₈W₄₈O₁₈₄)]36-occurs in the 1.5-4.7 pH range and corresponds to a fast and quantitative condensation process. Furthermore, 31P NMR titrations in solution reveal formation of the "monohandle" derivative [{Mo₄O₄S ₄(H₂O)₃(OH)₂}(P₈W ₄₈O₁₈₄)]38-as an intermediate prior to formation of the "bishandle" derivatives. Furthermore, the electrochemical behavior of [{Mo₄O₄S₄(H₂O)₃(OH) ₂}₂(P₈W₄₈O₁₈₄)]36-was studied in aqueous medium and compared with the parent anion P₈W ₄₈.

Full text of DOI:10.1021/ic202346r

Article
pubs.acs.org/IC
Oxothiomolybdenum Derivatives of the Superlacunary Crown
Heteropolyanion {P W }: Structure of [K {Mo O S (H O) (OH) } -
8
48
4
4 4 4
2
3
2 2
3
0−
(
WO )(P W O )] and Studies in Solution
2 8 48 184
†
,‡
,†
†
†
́ ́ ̂
Vladimir S. Korenev, Sebastien Floquet,* Jerome Marrot, Mohamed Haouas,
†
†
‡
‡
Israel-Martyr Mbomekalle, Francis Taulelle, Maxim N. Sokolov, Vladimir P. Fedin,
̈
́
,†
and Emmanuel Cadot*
†
Institut Lavoisier de Versailles, UMR 8180, University of Versailles, 45 avenue des Etats-Unis, 78035 Versailles, France
Nikolaev Institute of Inorganic Chemistry, Siberian Branch of the Russian Academy of Science, Novosibirsk 630090, Russia
‡
*
S Supporting Information
3
3−
ABSTRACT: Reaction of the cyclic lacunary [H P W O ]
7
8
48 184
2
+
anion (noted P W ) with the [Mo S O (H O) ] oxothioca-
8
48
2
2
2
2
6
tion led to two compounds, namely, [K {Mo O S (H O) -
4
4
4
4
2
3
3
0−
(
[
OH) } (WO )(P W O184)]
(denoted 1) and
2
2
2
8
48
3
6−
{Mo O S (H O) (OH) } (P W O184)]
(denoted 2),
4
4
4
2
3
2
2
8
48
which were characterized in the solid state and solution. In
the solid state, the structure of [K {Mo O S (H O) (OH) } -
(
4
4
4
4
2
3
2 2
3
0−
WO )(P W O184)] reveals the presence of two disordered
Mo O S (H O) (OH) } “handles” connected on both sides
2 8 48
2
+
{
4
4
4
2
3
2
of the P W ring. Such a disorder is consistent with the pres-
8
48
2+
ence of two geometrical isomers where the relative disposition of the two {Mo O S (H O) (OH) } handles are arranged in a
4
4
4
2
3
2
perpendicular or parallel mode. Such an interpretation is fully supported by 31P and W NMR solution studies. The relative
183
+
+
stability of both geometrical isomers appears to be dependent upon the nature of the internal alkali cations, i.e., Na vs K , and
2
+
increased lability of the two {Mo O S (H O) (OH) } handles, compared to the oxo analogous, was clearly identified by
4
4
4
2
3
2
3
1
183
significant broadening of the P and W NMR lines. Solution studies carried out by UV−vis spectroscopy showed that
36−
formation of the adduct [{Mo O S (H O) (OH) } (P W O )] occurs in the 1.5−4.7 pH range and corresponds to a fast
4
4
4
2
3
2
2
8
48 184
3
1
and quantitative condensation process. Furthermore, P NMR titrations in solution reveal formation of the “monohandle”
38−
derivative [{Mo O S (H O) (OH) }(P W O184)]
as an intermediate prior to formation of the “bishandle” derivatives.
4
4
4
2
3
2
8
48
36−
Furthermore, the electrochemical behavior of [{Mo O S (H O) (OH) } (P W O184)] was studied in aqueous medium and
4
4
4
2
3
2
2
8
48
compared with the parent anion P W .
8
48
2
4,25
INTRODUCTION
Current development in the chemistry of polyoxometalates
organometallic moieties such as [Ru(p-cymene)],
and even
■
2
0,26,27
oxometallic clusters.
In this way, combinations with the dinuclear oxothiocation
[Mo O S ] with P W should be promising to give new
insights about the coordination/condensation properties of this
large macrocyclic ion. P W represents a stable and an electro-
active component which could be viewed as an electron storage
while the {Mo O S }-based materials are able to
promote electrocatalysis, such as proton reduction into hydro-
Thus, some synergistic effects could be expected from
combinations of both components within a single complex unit.
1
(
POMs) is widely driven by potential applications in catalysis,
2+
2
3,4
medicine, magnetism, materials sciences, and nanotech-
2 2 2 8 48
5
−8
nology.
Accumulation of transition metal centers in a single
nanoscopic inorganic cluster can play a crucial role for catalytic
and electrocatalytic properties. Selective condensation of catio-
nic metallocomplex units on nucleophilic lacunary precursors
corresponds to a useful approach to synthezise high-nuclearity
POMs. In such a context, vacant polyoxotungstates constitute
ideal candidates because they exhibit remarkable rigidity and
robustness. In this field, the cyclic superlacunary anionic POM
8
48
16−18,28
unit,
2 2 2
29−31
gen.
2+
The dinuclear cation [Mo O S (H O) ] exhibits coordina-
tion requirements restricted to three available sites on each
molybdenum center, thus allowing one to develop predictable
2
2
2
2
6
3
3−
[
H P W O184]
(P W ) isolated in 1985 by Contant and
7
8
48
8 48
9
Teze
́
represents an attractive system which exhibits a high pro-
POM−{Mo O S } based coordination chemistry. The oxothio-
pensity to include various metallic cations in its large anionic
pocket. Actually, the host−guest chemistry based on the super-
lacunary P W was investigated and widely demonstrated
2
2 2
cation has been proved to react with basic lacunary POMs
32
33
34
35
36
37
such as PW9, AsW9, SiW , PW , PW , P W , and
10
10
11
2
15
8
48
through structural characterization of a large series of com-
1
0−17
plexes including various metallic cations such as copper,
Received: October 30, 2011
Published: January 23, 2012
1
8
19,20
20
20−22
23
iron, cobalt,
nickel, manganese,
or lanthanides,
©
2012 American Chemical Society
2349
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358

Inorganic Chemistry
Article
3
8
P2W17 to give species ranging from monomeric to tetra-
respectively. Spectra with reasonable signal-to-noise ratios were ob-
tained after 48 transients at a repetition time of 80 s. Chemical shifts
were calibrated relative to (NH )H PO as an external standard (set to
modular assembly.
Herein we report on the synthesis and characterization of
combinations between the P W48 and the [Mo O S ]
moieties. The synthesis and structure of the addition com-
4
2
4
31
δ = 0 ppm). The experimental errors of the P chemical shifts were
2
+
183
8
2
2 2
estimated to be ±0.4 ppm. Solution W NMR (20.833 MHz) spectra
were recorded at 26 °C in 10 mm o.d. tubes on a Bruker Avance-500
FT-NMR spectrometer equipped with a Bruker BBO low-frequency
tunable probe and Bruker Topspin data-acquisition software. These
36−
pounds based on the [{Mo O S (H O) (OH) } (P W O )]
4
4
4
2
3
2 2
8
48 184
architecture are described and discussed. Particular attention
is focused on the study of the association process in solution,
which highlights formation of an intermediate species
spectra were calibrated to an external secondary reference of SiW12
−
D O solution (δ = −103.8 ppm). The peaks were decomposed by
2
3
8−
least-squares computer fitting assuming pure Lorentzian shape for the
solution spectra and combined Gaussian and Lorentzian broadening
function for solid spectra; the chemical shift, half width, and intensity
of each peak were then estimated. The composite curves reproduced
well the experimental spectra.
[
{Mo O S (OH) (H O) }P W ] prior to formation of two
4 4 4 2 2 3 8 48
saturated geometric isomers, denoted para and perp. Further-
more, electrochemical data of the {Mo O S }-containing P W
compound are also given.
2
2
2
8
48
Electrochemistry Studies. Purified water was used throughout. It
was obtained by passing water through a RiOs 8 unit followed by a
Millipore-Q Academic purification set. All reagents were of high-purity
grade and used as purchased without further purification. The com-
positions of the various media were as follows: for pH 1, 2, and 3 0.5 M
EXPERIMENTAL SECTION
■
Physical Methods. Water content was determined by thermal
gravimetric analysis (tga7, Perkin-Elmer). Infrared spectra were
recorded on a Magna 550 Nicolet spectrophotometer in KBr pellets.
Elemental analyses were performed by the Service central d’analyze du
CNRS, Vernaison, France and by the Service d’analyze du CNRS, ICSN,
Gif sur Yvette, France. Energy-dispersive X-ray spectroscopy (EDX)
measurements were performed on a JEOL JSM 5800LV apparatus.
Spectrophotometric studies were carried out on a Perkin-Elmer
Lambda 19 spectrophotometer. Stability studies were carried out with
Li
SO + H SO ; for pH 4 and 5 1.0 M CH COOLi + CH COOH.
2
4 2 4 3 3
The polyanion concentration was 0.2 mM. All cyclic voltammograms
−1
were recorded at 10 mV s scan rate unless otherwise stated. Solutions
were deaerated thoroughly for at least 30 min with pure argon and
kept under a positive pressure of this gas during the experiments. The
source, mounting, and polishing of the glassy carbon (GC, Carbone
3
9
0
.1 mM solutions of the relevant polyanion. Matched 1 cm optical
path length quartz cuvettes were used. The compositions of the
various media were as follows: for pH 1, 2, and 3 0.5 M Li SO +
Lorraine, France) electrodes have been described. The glassy carbon
samples had a diameter of 3 mm. Controlled potential coulometry
experiments were performed with a large surface area glassy carbon
plate as the working electrode. The auxiliary electrode was a Pt plate
placed within a fritted-glass isolation chamber. Potentials are quoted
against a saturated calomel electrode (SCE). The electrochemical
setup was an EG&G 273 A potentiostat controlled by a PC with the
M270 software. All experiments were performed at room temperature.
X-ray Diffraction Studies. The structure of KLi-1 was obtained
from single crystals of K Li H [K {Mo O S (H O) (OH) } (WO )-
2
4
H SO ; for pH 4 and 5 1.0 M CH COOLi + CH COOH. Spectro-
2
4
3
3
photometric titrations were performed at room temperature in quartz
cells of 1 mm path length. Stock solutions (2.25 × 10 M) of P W in
−
3
8
48
−3
1
[
(
M LiCl at pH = 1.5, 3, or 4.7 and 2.25 × 10 M stock solutions of
2+
−4
Mo O S ] (4.5 × 10 M of K2−x(NMe ) [I Mo O S (OH) -
2 2 2 4 x 2 10 10 10 10
H O) ]·20H O) (with x between 0 and 0.5) in 1 M LiCl at
2
5
2
pH = 1.5, 3, or 4.7 were prepared by dissolution of P W or
8
48
20
6
4
4
4
4
4
2
3
2
2
2
K2−x(NMe ) [I Mo O S (OH) (H O) ]·20H O in 1 M LiCl
(P W O )]·95H O. A parallelepipedic red single crystal was
4
x
2
10 10 10
10
2
5
2
8 48 184 2
followed by adjustment of the pH, 1.5 and 3, with HCl. The batch
mounted in a capillary. Dimensions of the crystal and cell parameters
are given in Table 1. X-ray intensity data were collected at 293 K on a
Bruker-Nonius X8-APEX2 CCD area detector diffractometer using
Mo Kα radiation (λ = 0.71073 Å). Data reduction was accomplished
using SAINT V7.03 (APEX2 version 1.0-8; Bruker AXS: Madison, WI,
2003). The substantial redundancy in data allowed a semiempirical
absorption correction (SADABS V2.10) to be applied on the basis of
multiple measurements of equivalent reflections. The structure was
solved by direct methods, developed by successive difference Fourier
2+
method was used to prepare samples with ratios [Mo O S ] /P W
2
2
2
8
48
ranging from 0 to 4 in 1 M LiCl at pH = 1.5 and 3. The concentrations
2+
of the [Mo O S ] chromophore were maintained constant by adjusting
2
2 2
the total volumes with 1 M LiCl adjusted at pH = 1.5 or 3. The samples
were left to equilibrate overnight at room temperature before taking
readings. A similar procedure was followed for pH = 4.7 except that
the solutions were prepared in CH COOH/CH COOLi buffer
3
3
(
1M/1M).
NMR Measurements. Solution 31P NMR measurements were
2
syntheses, and refined by full-matrix least-squares on all F data using
performed on a Bruker Avance 300 instrument operating at 121.5 MHz
in 5 mm o.d. tubes. Chemical shifts were referenced to 85% H PO
SHELXTL V6.12 (SHELXTL version 6.12; Bruker AXS: Madison,
WI, 2001). Heavy atoms were initially located by direct methods. The
remaining non-hydrogen atoms were located from Fourier differences
and refined with anisotropic thermal parameters. All atoms were aniso-
tropically refined except for some disordered free water molecules.
Figures 1−4 were generated by Diamond version 3.2g (copyright
Crystal Impact GbR). The crystal structure may be obtained from the
Fachinformationszentrum Karlsruhe, 76344, Eggenstein-Leopoldshafen,
Germany (crystaldata@fiz-karlsruhe.de) on quoting depository number
423504.
3
4
solution. 31P NMR titrations were performed at pH = 1, 2, and 3
−
3
as follows: a 4.5 × 10 M stock solution of P W in 1 M LiCl at
8
48
pH = 1, 2, or 3 was obtained by dissolving solid P W in 1 M LiCl and
8
48
adjusting the required pH with HCl. To increase the solubility, the
major part of the potassium countercations associated to P W were
8
48
replaced by sodium through addition of NaBPh (24 equiv per P W
4
8
48
units) and removing the resulting white precipitate of KBPh .
4
−2
2+
Alternatively, 4.5 × 10 M stock solutions of [Mo O S ] in 1 M
LiCl at pH = 1, 2, or 3 were prepared without addition of NaBPh as
described above. Samples with ratios [Mo O S ] /P W ranging
2
2 2
4
Syntheses. Precursors K2−x(NMe ) [I Mo O S (OH) -
4
x
2
10 10 10
10
2
+
2
2
2
8
48
(H O) ]·20H O (x < 0.5) and K Li H [P W O ]·92H O
2 5 2 28 5 7 8 48 184 2
from 0 to 4 were prepared by mixing required amounts of Mo O S
(P W ) were prepared as described in the literature and characterized
8 48
9,40
2
2
2
solutions (0−0.4 mL) with 1 mL of the P W stock solution under
by routine methods (FT-IR, NMR, EDX, TGA).
the NMR data are given in Tables 2 and 3.
More details on
8
48
stirring followed by addition of 0.1 mL of D O. The samples were left
2
to equilibrate under stirring overnight at room temperature. Solid-state
K Li H [K {Mo O S (OH) (H O) } (WO )(P W O )]·95H O,
20 6 4 4 4 4 4 2 2 3 2 2 8 48 184 2
3
1
P NMR spectra were measured at room temperature under magic
KLi-1. P W (1 g, 0.068 mmol) and LiCl (0.5 g, 11.765 mmol) were
8 48
angle spinning (MAS) condition. NMR spectra were obtained on a
Bruker Avance-500 spectrometer equipped with a MAS accessory
operating at 202.456 MHz. A zirconium oxide cylindrical-type rotor
was used. Spectra were recorded with a single 45° pulse (2.4 μs) at a
spinning rate of about 30 kHz (no spinning side bands were
observed). The spectral width and data points were 15 kHz and 2 k,
dissolved in 25 mL of H O. The pH was adjusted to 2 with a few
2
drops of 6 M HCl. The solution was heated to 55−60 °C, and K2−x-
(NMe ) [I Mo O S (OH) (H O) ]·20H O (0.128 g, 0.056 mmol)
4
x
2
10 10 10
10
2
5
2
was added. The solution color changed from colorless to red. The
mixture was heated at 55−60 °C about 4 h. The resulting solution was
covered with Parafilm, and crystals were obtained within a few days.
2
350
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358

Inorganic Chemistry
Article
Table 1. Relevant Crystallographic Data for 1
1
formula
H210K Li Mo O P S W
24
6
8
299
8
8
49
−
1
M, g·mol
T, K
16 256.13
293(2)
wavelength/Å
cryst size (mm)
cryst syst
space group
a, Å
0.71073
0.16 × 0.10 × 0.06
tetragonal
I4/mmm
26.7803(11)
26.7803(11)
21.4115(10)
90
b, Å
c, Å
α, deg
β, deg
90
γ, deg
90
V, Å3
15356.0(11)
2
Figure 2. Top and side views of the coordination of the [Mo O S -
Z
4
4 4
2+
−
3
(OH) (H O) ] handles and the potassium cations with the {P W }
2
2
3
8
48
Dcalcd/g cm
μ/mm−1
3.516
moiety.
19.120
θ range for data collection
F(000)
1.08−30°
13 128
completeness to θmax/%
limiting indices
99.9
−37 ≤ h ≤ 37
−
−
33 ≤ k ≤ 37
30 ≤ l ≤ 28
data/restraints/params
6135/0/261
R(int)
0.0619
2
goodness-of-fit on F
1.191
a
final R indices [I > 2σ(I)]
R1 = 0.0684
b
wR2 = 0.1849
largest diff. peak and hole/e·A−
3
4.000 and −6.409
a
b
2
2 2
R1 = (∑|F | − |F |)/(∑|F |). Rw = ((∑w(Fo − F ) )/
o
c
c
c
2
2
1/2
2
2
o
2
(
∑w(F ) )) , 1/w = σ F + (aP) + bP.
o
Figure 1. (a) Representation of the structural disorder involving two
2
+
{
Mo S (OH) (OH ) } grafted on both sides of the {P W } surface
4 4 2 2 3 8 48
Figure 3. Top and side views of the parallel isomer para-
as the superimposition of two geometrical isomers with perpendicular
b) and parallel disposition (c). Color code: gray sticks = P W
32−
[
K P W O (Mo O S (OH) (H O) ) ]
of 1 in ball-and-stick
4
8
48 184
4
4
4
2
2
3 2
(
8
48
representation. Additional disordered tungstate groups are ommited.
Phosphates are depicted in pink polyhedra, tungsten in blue,
molybdenum in green, potassium in gray, oxygen in red, and sulfur
in yellow.
framework; orange sticks = {Mo S (OH) (OH ) } basket handle;
4
4
2
2 3
green spheres = disordered potassium cations.
Yield: ca. 0.6 g of red crystals (54%). IR/cm−1 (KBr pellet): 1138(m),
1
7
080(m), 1013(w), 956(m), 932(s), 910(s), 817(s, br), 747(s, br),
09(s, br), 666(sh), 571(m), 517(m), 459(w). TGA: An average loss
calculated for K Li H [K {Mo O S (OH) (H O) } (WO )-
2
0
6
4
4
4
4
4
2
2
3
2
2
(
P W O )]·95H O (found): W/Mo = 6.12 (6.31); Mo/S = 1.00
8 48 184 2
of ca. 100 water molecules between room temperature and 200 °C
corresponding to crystallization and coordinated water molecules.
Anal. Calcd for K Li H [K {Mo O S (OH) (H O) } (WO )-
(1.12); W/K = 2.04 (2.03); W/P = 6.12 (6.38).
K Li H [{Mo O S (OH) (H O) } (P W O )]·90H O, KLi-2.
2
6
2
8
4
4
4
2
2
3 2
8
48 184
2
2
0
6
4
4
4
4
4
2
2
3
2
2
P8W48 (3 g, 0.203 mmol) and LiCl (3.4 g, 80 mmol) were dissolved
−1
(
P W O )]·95H O (M = 16 256.1 g mol ): K, 5.77; Li, 0.26; P,
8 48 184 2
in 80 mL of 0.1 M HCl. NaBPh
added to remove potassium cations as a white precipitate KBPh . K
4
(1.6 g, 4.678 mmol, ∼23 equiv) was
1
1
.52; Mo, 4.72; W, 55.41; H, 1.30; S, 1.58. Found: K, 5.49; Li, 0.23; P,
.60; Mo, 4.76; W, 54.26; H, 1.10; S, 1.82. EDX atomic ratios
-
4
2−x
(NMe ) [I Mo O S (OH) (H O) ]·20H O (0.385 g, 0.167 mmol)
4
x
2
10 10 10
10
2
5
2
2
351
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358

Inorganic Chemistry
Article
Table 3. 183W NMR Data in LiCl (1 M) Compared to
Literature Data
chemical shifts,
ppm
compounds
P8W48 from ref 27
−190.1
−191.0
−
207.6
perp-[K (Mo O ) (Mo O (H O) )(P W O )]25−
VI
V
−137.1
8
2
2
4
10
2
3
8
48 184
from ref 27
−190.5
−193.0
−205.7
−215.6
−304.4
NaK-2
−165.0
−174.7
−197.1
−201.1
−205.9
−211.3
−222.4
Na-2
−196.3
−200.9
−221.8
7
58(s, br), 712(s, br), 668(sh), 569(m), 517(m), 458(w). TGA: Loss
of ca. 95 water molecules between room temperature and 200 °C
corresponding to free and coordinated water molecules. Anal. Calcd
for K Li H [{Mo O S (OH) (H O) } (P W O )]·90H O (M =
Figure 4. Top and side views of the perpendicular isomer perp-
32−
[
K P W O (Mo O S (OH) (H O) ) ]
in 1 in ball-and-stick
4
8
48 184
4
4
4
2
2
3
2
26
2
8
4
4
4
2
2
3
2
8
48 184
2
−1
representation. Additional disordered tungstate groups are ommited.
Phosphates are depicted in pink polyhedra, tungsten in blue,
molybdenum in green, potassium in gray, oxygen in red, and sulfur
in yellow.
16004.1 g mol ): K, 6.35; Li, 0.09; P, 1.55; Mo, 4.80; W, 55.14; H,
1.28; S 1.60. Found: K, 6.16; Li, 0.08; P, 1.64; Mo, 4.95; W, 53.93; H,
0.98; S, 1.79. EDX atomic ratio calculated for K Li H [{Mo O S -
26
2
8
4
4 4
(OH) (H O) } (P W O )]·90H O (found): W/Mo = 6.0 (6.03);
2
2
3
2
8
48 184
2
Mo/S = 1 (0.96); W/K = 1.85 (1.73); W/P = 6.0 (6.25).
Table 2. 31P NMR Data in LiCl (1 M)
Na K LiH [{Mo O S (OH) (H O) } (P W O )]·125H O, NaK-2.
25
2
8
4
4
4
2
2
3 2
8
48 184
2
The beginning of the procedure is similar to that described above
for compound KLi-2. The red resulting solution was stirred for 5 h;
then NaCl (10 g, 171 mmol) was added to give a red crystalline
precipitate. After 12 h at 4−6 °C, crystalline material was collected by
filtration, washed with water, ethanol, and ether, and dried in air. Yield:
ca. 0.755 g of the crystalline mixed Na/K/Li salt denoted NaK-2 (yield
compounds
signals/relative integration/bandwidth
P8W48
−6.56 ppm/1/3 Hz
−7.54 ppm/0.26/3 Hz
KLi-2
−8.18 ppm/0.53/3 Hz
−8.92 ppm/0.21/12 Hz
31
2
3%). P NMR in aqueous molar LiCl/ppm: −7.31, −8.35, −9.42.
NaK-2
−7.31 ppm/0.37/3 Hz
−1
IR/cm (KBr pellet): 1140(m), 1083(m), 1020(w), 961(s), 939(s),
−8.35 ppm/0.30/56 Hz
9
5
15(s), 841(sh, br), 780(s, br), 714 (s, br), 661(sh, br), 570(m),
12(m), 462(w). TGA: Loss of ca. 130 water molecules between room
−9.42 ppm/0.33/12 Hz
NaK-2 + 5 equiv of KCl
NaK-2 + 10 equiv of KCl
Na-2
−7.31 ppm/0.37/3 Hz
temperature and 200 °C corresponding to free and coordinated water
−7.92 ppm/0.30/14 Hz
molecules. Anal. Calcd for Na K LiH [{Mo O S (OH) (H O) } -
2
5
2
8
4
4
4
2
2
3 2
−9.02 ppm/0.33/21 Hz
−1
(
P W O )]·125H O (M = 16264.1 g mol ): Na, 3.53; K, 0.48; Li,
8 48 184 2
−7.31 ppm/0.37/3 Hz
0.04; P, 1.52; Mo, 4.72; W, 54.26; H, 1.70; S, 1.58. Found: Na, 3.27; K,
0.45; Li, 0.04; P, 1.35; Mo, 4.19; W, 54.00; H, 1.35; S, 1.60. EDX
atomic ratio calculated for Na K LiH [{Mo O S (OH) (H O) } -
−7.95 ppm/0.30/10 Hz
−8.97 ppm/0.33/23 Hz
25
2
8
4
4
4
2
2
3 2
−7.36 ppm/ca. 0.06/35 Hz
(P W O )]·125H O (found): W/Mo = 6.00 (6.10); Mo/S = 1.00
8 48 184 2
(1.03); W/P = 6.00 (6.13); W/Na = 1.92 (2.08); W/K = 12.0 (11.18).
−8.15 ppm/0.90/55 Hz
Na H [{Mo O S (OH) (H O) } (P W O )]·130H O·∼8 NaCl,
−9.29 ppm/ca. 0.04/60 Hz
28
8
4
4
4
2
2
3 2
8
48 184
2
Na-2. The beginning of the procedure is similar to that described
above for the previous compound. The dark red-colored filtrate col-
lected after precipitation of NaK-2 was allowed to stand for crystal-
lization in air. After several days, red crystalline compound mixed with
colorless crystals of NaCl was obtained. The red crystals denoted Na-2
were separated mechanically from NaCl, washed with ethanol and
Na-2 + 10 equiv of KCl
−7.29 ppm/ca. 0.06/16 Hz
7.86 ppm/0.90/12 Hz
ca. −8.95 ppm, very broad
−
was added, provoking color change from colorless to dark red. The
limpid mixture was stirred at room temperature. After 6 h, KCl (2 g,
6.846 mmol) was added to give a red-orange precipitate, which was
collected by filtration, washed with water, ethanol, and ether, and dried
in air. Yield: ca. 2.2 g of crude potassium salt (66%). P NMR in
31
ether, and dried in air. Yield: 0.67 g (∼21%). P NMR in aqueous
−1
2
molar LiCl/ppm: −7.36, −8.15, −9.29 (very broad). IR/cm (KBr
pellet): 1141(m), 1084(m), 1020(w), 961(s), 937(s), 915(s), 828(sh,
br), 770(s, br), 713(s, br), 668(sh, br), 571(m), 523(m), 461(w).
TGA: Loss of ca. 135 water molecules between room temperature
and 200 °C corresponding to free and coordinated water molecules.
31
−1
aqueous molar LiCl/ppm: −7.54, −8.18, −8.92. IR/cm (KBr pellet):
140(m), 1081(m), 1015(w), 956(m), 933(s), 912(s), 819(s, br),
1
2
352
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358

Inorganic Chemistry
Article
EDX atomic ratio calculated for Na H [{Mo O S (OH) (H O) } -
equivalent sites (SOF = 50%) which span two neighboring
28
8
4
4
4
2
2
3 2
(
1
P W O )]·130H O·∼8 NaCl (found): W/Mo = 6.0 (6.12); Mo/S =
8
48 184
2
P2W12 subunits with the usual K−O bonds (2.736−2.795 Å)
(1.03); P/Mo = 1 (1.09); W/Na = 1.33 (1.37); absence or only traces
(
see Figures 3 and 4). In addition, a weak electronic density was
of potassium.
found very close to the peripheral Mo atom and matches with
that corresponding to 1/8 tungsten atoms. Such a result is
consistent with previous results which show that these one-
eighth equivalent positions are able to coordinate cis dioxo
RESULTS AND DISCUSSION
■
Structure. Compound 1 crystallizes as a mixed potassium−
lithium salt in the tetragonal space group I4/mmm. The mole-
cular arrangement in 1 corresponds to a P W macrocyclic ring
2
0,22,24,25,27
metallic cation {MO } with M = Mo, W, or V.
This
2
W atom is bound to two oxygen atoms of two adjacent P W
2
12
8
48
units (W···O = 1.754 Å) and two terminal cis oxygen atoms
W···O = 1.781 and 1.901 Å), which complete the distorted
capped on both sides by two disordered [Mo O S (OH) -
4
4
4
2
2+
(
(
H O) ] oxothiomolybdenum clusters (Figure 1). On each
side, the [Mo O S (OH) (H O) ] hemicycle appears distrib-
2 3
2
+
2
+
tetrahedral environment. Obviously, the {WO2} group is
coordinated on the four coordination sites remaining free within
the para and perp isomers (see Figure 3 and 4), giving in such
4
4
4
2
2
3
uted with a statisticall occupancy factor of 50% over two
equivalent positions related through a C symmetry axis. Such
4
2+
conditions one {WO2} group per isomer (para and perp).
disorder can be analyzed through two different options. The
first one consists in the presence of two positional isomers
located on the same crystallographic site and was characterized
by the relative disposition perpendicular (noted perp) or parallel
Synthesis of the Compounds. 31P NMR solution study
(
not shown) reveals that 1 is not a pure compound. As
2+
surmised, the presence of one {WO2} group distributed over
two isomers and disordered over four positions in each isomer
could produce several compounds. As invoked by Kortz et al.,
the additional cis dioxotungstic group is most likely formed
2
+
(
noted para) of the two [Mo O S (OH) (H O) ] (Figure 1)
4 4 4 2 2 3
groups. The second way could result from a positional disorder
of a single isomer (either perp or para). Overall, 31P NMR
studies in solution and the solid state performed on parent
compound 2 (see below) are consistent with the presence of two
anions and then corresponding to the perp and para isomers.
Then, the observed crystallographic disorder has to be properly
analyzed according to the first option, as depicted in Figure 1. In
the structure of 1, the tetranuclear “basket handle” core results
from the connection of two {Mo O S } groups through face-
during synthesis of the P W derivatives, which can require
8
48
2
3,25,20
prolonged heating.
tion process of P
able to react further on the P
NMR and UV−vis studies of the {Mo
Under such conditions, the degrada-
W48 results in release of oxo−tungstic species
8
3
1
W
48 precursor. Herein, the
}/P 48 systems
P
8
O
2
S
W
8
2
2
reveal that formation of the para and perp isomers is
quantitative and proceeds fast at room temperature. Then, no
heating is required to produce the addition compound.
Actually, prolonged heating used for the synthesis of KLi-1
allows only the increase of the crystal quality required for X-ray
diffraction. Optimized synthetic procedures, carried out at
room temperature, allows isolating perp/para isomers as
mixed K−Li (compound KLi-2) or K−Na−Li salts (compound
NaK-2).
2
2 2
shared octahedra, involving two bridging hydroxo groups and a
water molecule (Mo···Mo = 3.317 Å, Mo···OHbridging = 2.116 Å,
Mo···OH2 bridging = 2.536 Å). In addition, the coordination sphere
of the two peripheral Mo atoms of the tetranuclear arch is
completed by a terminal inward directed aquo ligand (Mo···OH =
2
2
.331 Å) to give octahedral Mo(V) centers. Such a tetranuclear
fragment was already observed in the supramolecular assembly
formed by combination of [Mo O S ] and P W
coordination compound [Mo S O (OH) (H O) (pba)] .
2
+
37
NMR Studies. Characterization of Compound 2. 31
P
NMR data of compounds KLi-2, NaK-2, and Na-2 are reported
in Table 2. The 31P NMR spectrum of mixed potassium−
lithium salt KLi-2 in 1 M LiCl, depicted in Figure 5b, consists
and in the
2
2
2
2
15
2−
41
4
4
4
2
2
3
The molecular structure of 1 closely ressembles the fully oxo
2
7
analogue previously described by Mu
̈
ller et al., which repre-
2+
sents the unique example where the oxo cation {Mo O }
2
4
exhibits a similar mode of junction as those usually observed for
the oxothio {Mo O S }-based molecules. Nevertheless, both
2
2 2
tetranuclear arrangements, oxo versus oxothio, differ slightly by
their protonation degree, although both complexes were synthetized
in the same pH range. The oxo derivative results in connections
between both dinuclear units through a double oxo bridge, giving
V
the neutral core [Mo O (H O) ], while such connec-
4
10
2
3
tions are ensured trough a double hydroxo bridge within the
oxothio derivative. Such a difference arises probably from the
softening of the Mo centers though the presence of the double
sulfido bridge, which weakens and stretches significantly the trans
Mo−O bonds. Such a Mo-softening effect found expres-
sion in differing self-condensation properties, which lead to
2+
42
exclusively cyclic linear oligomers with {Mo O S }
and com-
2
2 2
2+
43
pact stacked arrangements with {Mo O } . In these specific
2
4
cases the common mode of junction of dimolybdic units could
be due to a vectorial growth of the self-condensation process,
which is strongly directed by the coordination require-
ment of the P W macrocycle. The tetranuclear arches are
Figure 5. 31P NMR spectra of compounds P W (a), KLi-2 (b), NaK-
8
48
8
48
grafted on the anion surface through W−O−Mo bridges
2
(c), and Na-2 (d) in aqueous LiCl (1 M).
(
Mo···O = 2.148 Å) (see Figure 2). As commonly observed
9
,23,25−27
for P W derivatives,
four potassium cations are
of three resolved lines located at −7.54, −8.18, and −8.92 ppm.
8
48
found inside the cavity, statistically distributed over four pairs of
It differs significantly from that of the free P W precursor,
8
48
2
353
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358

Inorganic Chemistry
Article
which gives a single resonance at −6.56 ppm (see Figure 5a).
The three resonances are attributed to a mixture of the two
isomers, namely, para- and perp-[{Mo O S (OH) (H O) } -
KLi-2 (see Figure S1, Supporting Information). Such results
suggest that in the enriched Na salt NaK-2 the cavity encloses the
+
sodium cation, which is labile enough to be replaced by potassium.
The strongest affinity of the cavity for the potassium is related to
the size requirement of the eight sites which exhibit in the solid
4
4
4
2
2
3 2
36−
P W ] (Figures 3 and 4). The parallel arrangement should
8
48
3
1
retain a nominal D symmetry, giving the two P NMR signals
2h
9
of equal intensities observed at −7.54 and −8.92 ppm, while
the para isomer gives the single resonance at −8.18 ppm,
related to the four equivalent phosphate groups within the
D idealized symmetry. For both isomers, the {Mo O S } coordi-
state seven K−O distances in the 2.73(2)−2.79 Å range. From
these considerations it seems consistent to consider faster
exchange rates with the smallest cations such as sodium. The
simplest dynamics could be viewed as intramolecular hopping of
each inner cation within the four pairs of sites, where both sites are
separated by a short distance of 2.48(2) Å. Within the para
isomers these four pairs are composed of two equivalent sites,
leading to a strictly equal averaged distribution of the sodium
cations through a fast hopping. In the perp isomer, these sites
become inequivalent, which could result in a lower hopping
frequency of the sodium cation on the NMR time scale,
experimentally observed as a broad 31P NMR resonance. The
potassium-free compound Na-2 which correspond to the second
crop of crystallization in the synthesis of NaK-2 is characterized by
2d
2
2 2
nation of the P W surface is characterized by a significant
8
48
3
1
shielding at the P nuclei. In the para isomer, one pair of equi-
valent phosphates is located far from the two [Mo O S (OH) -
4
4
4
2
2+
(
H O) ] handles and then could correspond to the high-
2 3
frequency resonance at −7.54 ppm, while the remaining pair,
sited in the vicinity of both para-arranged [Mo O S (OH) -
4
4
4
2
2+
(
H O) ] handles, could be assigned to the low-frequency line
2 3
at −8.92 ppm. Interestingly, the low-frequency resonance at
−
8.92 ppm appears significantly broadened (Δν = 12 Hz)
1
/2
compared to the other (Δν = 3 Hz). Such an observation
1
/2
31
could be related to a mutual trans influence of the two parallel
very broad P NMR peaks which likely involves both isomers i.e.,
2
+
[
Mo O S (OH) (H O) ] handles, which spread along the
perp compound (δ = −8.15 ppm; Δν = 55 Hz) and para
4
4
4
2
2
3
1/2
W−O−W bridges. In agreement with such an attribution, the
species (δ = −7.3 ppm, Δν ≈ 35 Hz and −8.95 ppm; Δν
1
/2
1/2
31
four equivalent phosphates within the perp isomer are close to
≈ 60 Hz). The P NMR spectrum of Na-2 features typical
signals at the limit of the coalescence and then suggests that in
the absence of potassium ion both isomers are related through
a dynamic equibrium involving a decoordination/coordination
2
+
only one [Mo O S (OH) (H O) ] handle and far from the
4
4
4
2
2
3
second, thus giving a resonance observed at −8.18, about
31
halfway between −7.54 and −8.92 ppm. Besides, this P NMR
2
+
chemical shift appears in good agreement with that found
process of the two [Mo O S (OH) (H O) ] handles.
4 4 4 2 2 3
3
6−
31
for[(VO ) (P W O184)] (δ = −8.1 ppm) which exhibits a
Furthermore, quantitative analysis of the P NMR spectrum
indicates that Na-2 contains about 85% perp isomer. The
2
4
8
48
+
20
31
similar distribution of the four {VO } within the cavity. The
P
2
31
ratio of both isomers in KLi-2 calculated from integrated
NMR lines gives values around 55 ± 5% for the perp isomer vs
5 ± 5% for the para compound. Such an observation is consis-
P
NMR spectrum of an aged solution of Na-2 (1 month later)
does not show any significant changes in the isomeric perp/
para ratio (see Figure S2, Supporting Information), thereby
evidencing an equilibrium state. As previously mentioned,
addition of small amounts of KCl to a solution of Na-2 (5−10
equiv) provokes an important sharpening of the middle peak
from 55 to 12 Hz (see Figure S3, Supporting Information)
previously attributed to the perp isomer, but surprisingly, the
low-frequency resonance assigned to one phosphate group
within the para isomer disappears, probably due to a substantial
broadening, while the other resonance of the para species
remains unchanged. Furthermore, the ratio perp/(para + perp)
decreases slightly from 85% to 65%. These qualitative results
4
tent with a slightly unfavorable trans influence for the para isomer
with respect to the perp arrangement. The perp/para ratio does
not vary significantly neither with time nor with temperature in the
2
75−345 K range. Evidence of two isomers for such a system, i.e.,
{Mo O S }-containing P W , contrasts clearly with results ob-
2
2
2
8
48
tained by Mul
̈
ler et al. about the oxo derivative [K (MoO ) -
8 2 4
2
4− 27
(
Mo O (H O) ) (P W O )]
.
The latter gives only a
4
10
2
3 2
8
48 184
3
1
single sharp P NMR resonance, consistent with the presence of
27
the perpendicular arrangement as a single isomer.
The 31P NMR spectrum of the mixed sodium−potassium−
lithium salt NaK-2 in 1 M LiCl (Figure 5c) differs significantly,
revealing that chemical shifts and line widths are quite sensitive
to the nature of the embedded countercations (see Table 2). In
NaK-2, the separation of two signals corresponding to the para
isomer appears significantly more important (Δδ = 2.1 ppm)
while their line width remains unchanged. The 31P NMR
resonance of the perp isomer remarkably broadens from 3 Hz in
KLi-2 to 56 Hz in NaK-2, and the ratio perp/para is reversed to
give 0.3/0.7. This result suggests that the nature and distribution
of the enclosed countercations have a significant influence upon
the charge density by the coordinating oxygen atoms lining both
+
+
show that the nature of the counterion Na vs K has a great
influence upon (i) the relative stability of both iso-
mers and (ii) the dynamic exchange of the intramolecular cation
with a significant effect on the lability of the two [Mo O S -
4
4 4
2+
(OH) (H O) ] handles grafted on the P W surface.
2
2
3
8
48
In addition to the solution studies, compounds NaK-2 and
Na-2 were characterized by solid-state 31P NMR. P MAS
NMR spectra (see Figure S4, Supporting Information) exhibit a
large peak which results from the overlap of the three reso-
31
31
nances related to the presence of both isomers. P MAS NMR
of compound Na-2 reveals only one signal with symmetric
shape at −9.5 ppm which could be related to the resonance of
the perp isomer as the predominant species. Conversely, the
compound NaK-2 reveals a large asymmetric signal charac-
terized by a peak maximum at −8.3 ppm and a shoulder at
−9.7 ppm. Such an asymmetric shape could be due to the
presence of both isomers and can be analyzed as follows: these
two chemical shifts, i.e., −8.3 and −9.7 ppm, could correspond to
the two expected equal lines of the para isomer, while the addi-
tional contribution of the perp isomer could be found close to
−8.7 ppm. Deconvolution of both lines at −8.3 and −9.7 ppm
sides of the P W cavity. Furthermore, the variation of the 31
P
8
48
NMR line width from 3 to 56 Hz, observed for the perp-isomer,
indicates that dynamics is strongly influenced by the nature of the
+
+
counterion K versus Na and by the nature of the isomer. Several
pathways for dynamics can be postulated, such as cation exchange
from the cavity to the solvent or intramolecular exchange involving
the four pairs of sites identified by X-ray diffraction (see above).
Interestingly, addition of 5−10 equiv of KCl to a solution of NaK-
3
1
2
leads to a sharpening of the P NMR resonance of the perp
31
isomer and allows recovering an overall P spectrum close to
2
354
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358

Inorganic Chemistry
Article
gives an integrated intensity ratio of about 70/30, which leads to
a proportion of about 40% for the perp isomer and 60% for the
para species. Such values appear in agreement with the isomeric
distribution found in solution (see Table 2).
cisoid or transoid species. In the present case, this behavior
allows direct control of the coordination/decoordination pro-
cesses induced by the inner countercations.
3
1
31
P NMR Titrations. Batch P NMR titrations of P W by
8
48
2+
[
Mo O S ] in aqueous medium were performed at pH = 1, 2,
Compounds NaK-2 and Na-2 have been also characterized
2 2 2
by their 138W NMR spectra, shown in Figures 6 and S5,
3, and 4.7 after exchanging of the main part of potassium
cations for sodium to increase solubility. Figure 7 shows the
Figure 6. 183W NMR spectra of compounds Na-2 (a) and NaK-2 (b)
in aqueous LiCl (1 M).
Supporting Information. 183W NMR data are given in Table 3.
Surprisingly, 1 W NMR spectra of NaK-2 and Na-2 are very
83
183
different. The spectrum of NaK-2 exhibits seven W NMR
broad resonances, while that of Na-2 reveals only three equal
broad signals at −196.3, −200.9, and −221.8 ppm. Further-
1
83
more, it is noteworthy that W NMR spectra of both NaK-2
and Na-2 contrast clearly with that obtained from the oxo
V I
V
derivative perp-[K (Mo O ) (Mo O1 0 (H O) )-
8
2
4
4
2
3
24−
183
(
P W O184)] , where the six sharp W NMR lines were
8 48
fully consistent with the structure found by X-ray crystallog-
2
7
raphy. In the case of the {Mo O S }-containing P W
2
2
2
8
48
derivatives, the presence of the two isomers perp and para
involved in dynamic processes have to be considered. The pres-
ence of three 1 W NMR equal resonances for Na-2 indicates
an apparent D4h symmetry for the P W skeleton, while lower
83
8
48
symmetries D2d and D2h are expected for the frozen isomers
perp and para, respectively. Then the D4h symmetry must
necessarily correspond to an averaged symmetry, arising from a
dynamic process where the para and perp species are the limit-
ing structures. Such a process can be viewed as the fast wheel-
Figure 7. 31P NMR titrations of P W by [Mo O S ] in aqueous
2+
8
48
2
2 2
LiCl medium at pH = 3. Ratio R corresponds to the ratio
2+
[
[
Mo O S ] /P W . P W
(□), intermediate compound
48
2
2
2
8
48
8
38−
{Mo O S (OH) (H O) }P W ]
(H O) } P W ]
P W ] (●).
(▲), para-[{Mo O S (OH) -
4
4
4
2
2
3
8
48
4
4
4
2
2
+
ing of both [Mo O S (OH) (H O) ] handles on opposite
36−
4
4
4
2
2
3
(○), and perp-[{Mo O S (OH) (H O) } -
2
3
2
8
48
4 4 4 2 2 3 2
3
6−
faces of P W . Such a dynamic process was already postulated
8
48
8
48
3
1
from analysis of the P NMR data of Na-2 because the broad
signals attributed to the para and perp isomers were at the limit
results obtained at pH = 3, whereas the titrations carried out at
pH = 1, 2, and 4.7 are depicted in Figure S6, Supporting
Information. In all titration experiments addition of [Mo O S ]
to P W causes initially growing of two signals of the same in-
tensities in the −7.7 to −8.4 ppm range, whereas the signal of
free P W decreases. Intensities of these two signals are maxi-
mum for ratio [Mo O S ] /P W = 2. For higher ratios, these
two lines decrease gradually together for the benefit of the
signals corresponding to the saturated adducts, distributed as
1
83
of coalescence. The W NMR spectrum of NaK-2 appears
1
83
2+
more complex because the seven observed W NMR lines
should be the contribution of both isomers, although theo-
retically two sets of six resonances are expected. Actually, acci-
dental degeneracy can occur because in both isomers some
2
2 2
8
48
8
48
1
83
2+
tungsten atoms retain very close chemical environments.
W
2 2 2 8 48
and 31P NMR studies lead to the univocal conclusion that the
nature of the inner alkali cation governs to some extent the
interconversion rate between para/perp isomers and even
influences the isomeric ratio at the equilibrium. Both related
effects arise from very weak coordination bonds between the
two {Mo S (OH) (OH ) } “handles” and the P W macro-
2+
31
para and perp isomers. Up to ratio [Mo O S ] /P W = 4, P
2
2
2
8
48
NMR spectra do not change significantly (not shown) and con-
sist mainly of the three resonances attributed to the para and perp
isomers. As suggested by the stoichiometry and the feature of the P
31
4
4
2
2 3
8
48
cycle. The weakness of POM-{Mo O S } connections was
NMR spectra, the two signals found in the −7.7 to −8.4 ppm range
2
2 2
10−
already evidenced for dimeric assemblies [(PW ) (H Mo S O )]
have to be assigned to an intermediate, consistent with the
1
1 2
4
4
4
6
1
6− 36,38
38−
and [(P W ) (H Mo S O )]
,
which isomerize easily in
“monohandle” derivative [{Mo O S (OH) (H O) }P W ]
.
2
17 2
4
4
4
6
4
4
4
2
2
3
8
48
2
355
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358

Inorganic Chemistry
Article
2,34,36−38
3
In such conditions, the preferential formation of the para isomer is
characteristic of the red-brown thioderivatives of POMs
31
31
observed (para/(para + perp) ≈ 80%). From the P NMR
spectra and their interpretations the speciation curve can be
obtained and are shown in Figure 8. In summary, 31P NMR
since {Mo
2 2 2 8
O S }-based species and P W48 alone do not exhibit
significant absorptions in this region. Gradual addition of P
W
48
8
2+
to a constant quantity of [Mo O S ] leads to gradual in-
creasing of the absorbance in this region until a break point at
P W /[Mo O S ] = 1/4, consistent with the stoichiometry
2
2 2
2+
8
48
2
2 2
of the saturated complex (Figure 9b). Similar results are ob-
tained at pH = 1.5 and 4.7 (Figures S7 and S8, Supporting
Information), thus confirming binding of four oxothiocations
2
+
[
Mo O S ] per one P W in diluted solutions.
2 2 2 8 48
Stability and Electrochemistry Studies. Stability studies
in solution as a function of pH (from 1 to 5) were done by
UV−vis spectroscopy and cyclic voltammetry. The stability
criterion was reproducibility for spectra and cyclic voltammo-
grams (CVs) after several hours of running experiments. We
found that compounds P W , KLi-1, and Na-2 were stable
8
48
enough in sulfate buffer solution from pH 1 to 3 (0.5 M Li SO +
2
4
Figure 8. Speciation curve obtained for the 31P NMR titration per-
formed at pH = 3: P W (□), intermediate compound [{Mo O S -
H SO ) to be characterized by cyclic voltammetry. It was therefore
2
4
possible and easy to draw direct comparison between their
respective electrochemical behaviors in solution.
8
48
4
4 4
3
8−
(
OH) (H O) }P W ]
(▲), para-[{Mo O S (OH) (H O) } -
2
2
3
8
48
4
4
4
2
2
3 2
36−
36−
P W ] (○), and perp-[{Mo O S (OH) (H O) } P W ] (●).
8
48
4
4
4
2
2
3
2
8
48
Figure 10 compares Na-2 and P W CVs recorded under
8
48
the same conditions (0.5 M Li SO + H SO /pH = 2.0;
2
4
2
4
titrations allowed (i) evidencing formation of an intermediate
consistent to the monohandle derivative [{Mo O S (OH) -
4
4
4
2
38−
(
H O) }P W ] and (ii) identifying preferential formation of
2 3 8 48
36−
para-[{Mo O S (OH) (H O) } P W ] (80%) over the perp-
4
4
4
2
2
3 2
8
48
6−
3
[
{Mo O S (OH) (H O) } P W ] isomer (20%).
4
4
4
2
2
3 2
8
48
Spectrophotometric Studies. Batch spectrophotometric
2
+
titrations of a [Mo O S ] solution with P W were per-
2
2
2
8
48
formed in 1 M LiCl at pH = 1.5, 3, and 4.7. The results
obtained at pH = 3 are shown in Figure 9, whereas titrations at
Figure 10. Cyclic voltammograms of Na-2 (red) and P W (blue) in
8
48
−1
0
.5 M Li SO , pH = 2. Scan rate of 10 mV·s ; working electrode
2 4
glassy carbon. Potentials are quoted against SCE reference electrode.
POM concentrations = 0.2 mM.
−1
scan rate 10 mV s ). The CV of Na-2 is characterized by three
reduction steps located at −0.38, −0.47, and −0.61 V versus
SCE reference electrode. Compared to P W reduction peaks,
8
48
there is a potential shift of 80, 70, and 50 mV for the first,
second, and third reduction steps, respectively, at each time in
favor of P W reduction (see Table 4). These differences
8
48
Table 4. Reduction Peak Potentials, E , for the First and
pc
Second Tungsten Waves Determined by Cyclic Voltammetry
a
at pH 1, 2, and 3 for Compounds Na-2 and P W
8
48
2
+
Figure 9. UV−vis titration of [Mo O S ] by P W in aqueous LiCl
2
2
2
8
48
pH = 1
pH = 2
pH = 3
medium at pH = 1. Evolution of the resulting spectrum upon ad-
dition of increasing amounts of P W (a). Variation of the absorbance
8
48
Epc1
Epc2
Epc1
Epc2
Epc1
Epc2
measured at 470 nm as a function of the ratio R = P W /
8
48
Na-2
−0.29
−0.20
−0.40
−0.30
−0.38
−0.30
−0.47
−0.40
−0.45
−0.40
−0.53
−0.49
2+
[
Mo O S ] .
2 2 2
P8W48
a
−1
pH = 1.5 and 4.7 are shown in the Supporting Information
Figures S7 and S8). Absorptions in the 420−500 nm range are
Scan rate 10 mV·s , working electrode glassy carbon, reference
electrode SCE.
(
2
356
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358

Inorganic Chemistry
Article
observed at pH = 2 could suggest a higher basicity and/or a
CONCLUSION
This study was devoted to the reactivity of the superlacunary
■
higher formal negative charge for Na-2 as compared to P W48
8
under such conditions.
2+
P W with the oxothiocation [Mo O S ] in aqueous
8
48
2
2 2
Another characteristic that merited special study is the
number of electrons exchanged in each redox step. It is well
established that P W is a real electron sink (8 electrons for the
31
183
medium. Solution studies, carried out by P and W NMR
and UV−vis spectroscopy, demonstrate that coordination
8
48
of {Mo S O } on P W leads to an intermediate, corre-
44
2
2
2
8
48
two first waves and 8 for the third one). Controlled potential
coulometry experiments were carried out on a solution of Na-2
in 0.5 M Li SO + H SO /pH = 3.00. With the working elec-
sponding probably to the monohandle derivative prior to
formation of the saturated adducts. A structural model has been
established by X-ray diffraction study showing a structural
disorder explained by superimposition of both geometrical para
and perp isomers. Moreover, the influence of the countercations
2
4
2
4
trode potential set at −0.55 V vs SCE (first two reduction
steps) the electrical charge that flows through the solution
corresponded to an average of 8.3 ± 0.1 mol of electrons
consumed per mole of Na-2. This value is consistent with
+
+
K versus Na upon the equilibrium state and the dynamic
properties of the saturated bishandle derivatives was also
demonstrated qualitatively. The {Mo O S }-containing P W
48
reduction of the tungstic framework of P W . When the poten-
8
48
2
2
2
8
tial is set at −0.72 V vs SCE, i.e., on the third wave, the number
of electrons per mole of Na-2 is much higher than the expected
anions represent relevant chemical models to study the internal
distribution of cation in solution, crucial for understanding the
host−guest properties of the large macrocyclic anion P W .
16 mol. This difference is certainly due the electrocatalytic
8
48
reduction of protons that takes place here. This explains the
magnitude of this third reduction wave and its irreversible nature,
A quantitative study of the host−guest properties of the
{
Mo O S }-P W derivatives is under investigation using
2 2 2 8 48
as compared to the same wave for P W48 (Figures 10 and S9,
23
39
7
8
multinuclear NMR methods ( Na, K, Li).
Supporting Information). Indeed, one can list at least three
conditions that, when combined, would promote this phenom-
enon: the pH of the medium (3.00), the relative low potential
ASSOCIATED CONTENT
Supporting Information
P NMR spectra of NaK-2 in aqueous 1 M LiCl without
■
*
S
(
−0.72 V vs SCE), and the presence of two {Mo O S } entities
4
4
4
31
+
which in many cases are described as good electrocatalysts for H
reduction.
29−31
addition of KCl and after addition of 5 and 10 equiv of KCl
The presence of these two {Mo O S } entities is
31
4
4 4
(
1
Figure S1); P NMR spectra of compound Na-2 in aqueous
M LiCl recorded after 1 month compared to a spectrum
also highlighted by an irreversible oxidation wave that appears
above +0.65 V vs SCE at pH 3 (see Figure S10, Supporting
Information). For comparison, the CV of the molybdenum
recorded just after preparation of the NMR tube (Figure S2);
P NMR spectra of the sodium salt Na-2 in aqueous 1 M LiCl
31
2−
“
wheel” [Mo S O (OH) (C O )] recorded under the same
8 8 8 8 2 4
without addition of KCl and after addition of 10 equiv of KCl
conditions displays an irreversible wave with an oxidation peak
potential located at +0.88 V vs SCE (see Figure S10, Supporting
Information), which could be assigned to irreversible oxidation of
(
Figure S3); 31P MAS NMR spectra of compounds KLi-2,
1
83
3
NaK-2, and Na-2 in the solid state (Figure S4); simulated
NMR spectra of NaK-2 in aqueous 1 M LiCl (Figure S5);
NMR titrations of P W by [Mo O S ] in aqueous LiCl
W
1
V
VI
P
Mo into Mo .
2+
8
48
2
2 2
In aqueous solution, the redox behavior of Na-2 is pH
dependent, as seen in Figure 11. As the pH of the electrolyte
medium at pH = 2 and 4.7 (Figure S6); results of the UV−vis
2
+
titration of [Mo O S ] by P W in aqueous LiCl medium at
2
2
2
8
48
pH = 1.5 and 4.7 (Figures S7 and S8); cyclic voltammograms of
Na-2 and P W in 0.5 M Li SO + H SO pH 3 between −0.8
8
48
2
4
2
4
and +1.4 V (Figure S9); cyclic voltammograms of
2
−
[
Mo O S (OH) (C O )] in 0.5 M Li SO + H SO at pH
8 8 8 8 2 4 2 4 2 4
3
(Figure S10); X-ray crystallographic data in CIF format. This
material is available free of charge via the Internet at http://
pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
E-mail: sebastien.floquet@chimie.uvsq.fr (S.F.); cadot@
chimie.uvsq.fr (E.C.).
■
*
ACKNOWLEDGMENTS
We gratefully acknowledge the Centre National de la
Recherche Scientifique (CNRS) and the Minister
l’Education Nationale de l’Enseignement Super
■
̀
e de
ieur et de la
́
Figure 11. Cyclic voltammograms of Na-2 at different
Recherche (MENESR) for their financial support. S.F. is
grateful to Mr. Jimmy Mercier and Mr. Duc-Nam Luu for their
contributions to the experimental work. We are also grateful to
the GDRI Suprachem and the French Ministery of Foreign
Affairs for their support through the Ph.D. grant for V.S.K. This
work was also supported by the National Research Agency
(ANR) under contract POMEAH·ANR-08-JCJC-0097. The
work was supported by Russian Foundation for Basic Research
(Grant 09-03-93105).
pH values: pH 1 (black), 2 (red), and 5 (blue). Polyoxometalate
−1
concentration 0.2 mM; scan rate 10 mV·s ; working electrode: glassy
carbon; reference electrode SCE.
increases from 1 to 3, the reduction waves are shifted in the
direction of positive potentials but no noticeable change of
their shape is observed. Table 4 summarizes reduction peak
potential values for the first and second waves of Na-2 in 0.5 M
Li SO at pH 1, 2, and 3.
2
4
2
357
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358

Inorganic Chemistry
Article
(
29) Hijazi, A.; Kemmegne-Mbouguen, J. C.; Floquet, S.; Marrot, J.;
REFERENCES
■
Mayer, C. R.; Artero, V.; Cadot, E. Inorg. Chem. 2011, 50, 9031−9038.
(
1
1) Proust, A.; Thouvenot, R.; Gouzerh, P. Chem. Commun. 2008, 16,
(30) Duval, S.; Floquet, S.; Simonnet-Jegat, C.; Marrot, J.; Biboum,
́
837−1852.
R. N.; Keita, B.; Nadjo, L.; Haouas, M.; Taulelle, F.; Cadot, E. J. Am.
Chem. Soc. 2010, 132 (6), 2069−2077.
(
(
2) Ogata, A.; Yanagie, H.; Ishikawa, E.; Morishita, Y.; Mitsui, S.;
Yamashita, A.; Hasumi, K.; Takamoto, S.; Yamase, T.; Eriguchi, M. Br.
J. Cancer 2008, 98 (2), 399−409.
3) Kortz, U.; Muller, A.; van Slageren, J.; Schnack, J.; Dalal, N. S.;
̈
31) Keita, B.; Floquet, S.; Lemonnier, J. F.; Cadot, E.; Kachmar, A.;
́
ard, M.; Rohmer, M. M.; Nadjo, L. J. Phys. Chem. C 2008, 112 (4),
Ben
109−1114.
32) Bereau, V.; Cadot, E.; Bogge, H.; Mu
Inorg. Chem. 1999, 38 (25), 5803−5808.
33) Pilette, M.-A.; Marrot, J.; Secheresse, F.; Cadot, E. Inorg. Chim.
Acta 2010, 363, 4253−4261.
34) Cadot, E.; Bereau, V.; Marg, B.; Halut, S.; Sec
Chem. 1996, 35 (11), 3099−3106.
35) Cadot, E.; Bereau, V.; Secheresse, F. Inorg. Chim. Acta 1996, 252
1−2), 101−106.
́
36) Marrot, J.; Pilette, M. A.; Secheresse, F.; Cadot, E. Inorg. Chem.
(
1
(
Dressel, M. Coord. Chem. Rev. 2009, 253 (19−20), 2315−2327.
4) Mialane, P.; Dolbecq, A.; Secheresse, F. Chem. Commun. 2006,
3, 3477−3485.
5) Miras, H. N.; Cooper, G. J. T.; Long, D. L.; Bogge, H.; Mu
Streb, C.; Cronin, L. Science 2010, 327 (5961), 72−74.
6) Ritchie, C.; Cooper, G. J. T.; Song, Y. F.; Streb, C.; Yin, H. B.;
Parenty, A. D. C.; MacLaren, D. A.; Cronin, L. Nat. Chem. 2009, 1 (1),
7−52.
7) Pradeep, C. P.; Misdrahi, M. F.; Li, F. Y.; Zhang, J.; Xu, L.; Long,
D. L.; Liu, T. B.; Cronin, L. Angew. Chem., Int. Ed. 2009, 48 (44),
309−8313.
8) Rodriguez-Albelo, L. M.; Ruiz-Salvador, A. R.; Sampieri, A.; Lewis,
D. W.; Gomez, A.; Nohra, B.; Mialane, P.; Marrot, J.; Secheresse, F.;
́
̈
́
ller, A.; Secheresse, F.
(
3
(
́
(
́
̈
ller, A.;
(
́
́
heresse, F. Inorg.
(
(
́
́
4
(
(
(
2
003, 42 (11), 3609−3615.
8
(
(37) Cadot, E.; Pilette, M. A.; Marrot, M.; Secheresse, F. Angew.
́
Chem., Int. Ed. 2003, 42 (19), 2173−2176.
́
(38) Pilette, M.-A.; Floquet, S.; Marrot, J.; Cadot, E. Eur. J. Inorg.
Chem. 2011, 3523−3528.
Mellot-Draznieks, C.; Biboum, R. N.; Keita, B.; Nadjo, L.; Dolbecq, A.
J. Am. Chem. Soc. 2009, 131 (44), 16078−16087.
9) Contant, R.; Tez
10) Mal, S. S.; Bassil, B. S.; Ibrahim, M.; Nellutla, S.; van Tol, J.;
Dalal, N. S.; Fernandez, J. A.; Lopez, X.; Poblet, J. M.; Biboum, R. N.;
Kelta, B.; Kortz, U. Inorg. Chem. 2009, 48 (24), 11636−11645.
11) Chen, L. F.; Hu, J. C.; Mal, S. S.; Kortz, U.; Jaensch, H.; Mathys,
(39) Keita, B.; Nadjo, L. J. Electroanal. Chem. 1988, 243 (1), 87−103.
(
(
́
e,
́
A. Inorg. Chem. 1985, 24 (26), 4610−4614.
́
(40) Cadot, E.; Salignac, B.; Marrot, J.; Dolbecq, A.; Secheresse, F.
Chem. Commun. 2000, 4, 261−262.
(41) Lemonnier, J. F.; Floquet, S.; Marrot, J.; Cadot, E. J. Cluster Sci.
2006, 17 (2), 267−282.
(
(42) Lemonnier, J. F.; Duval, S.; Floquet, S.; Cadot, E. Isr. J. Chem.
2011, 51 (2), 290−302.
G.; Richards, R. M. Chem.Eur. J. 2009, 15 (30), 7490−7497.
12) Bao, Y. Y.; Bi, L. H.; Wu, L. X.; Mal, S. S.; Kortz, U. Langmuir
009, 25 (22), 13000−13006.
13) Liu, G.; Liu, T. B.; Mal, S. S.; Kortz, U. J. Am. Chem. Soc. 2006,
28 (31), 10103−10110.
14) Alam, M. S.; Dremov, V.; Muller, P.; Postnikov, A. V.; Mal, S. S.;
Hussain, F.; Kortz, U. Inorg. Chem. 2006, 45 (7), 2866−2872.
15) Mal, S. S.; Kortz, U. Angew. Chem., Int. Ed. 2005, 44 (24),
777−3780.
16) Jabbour, D.; Keita, B.; Nadjo, L.; Kortz, U.; Mal, S. S. Electro-
chem. Commun. 2005, 7 (8), 841−847.
17) Pichon, C.; Mialane, P.; Dolbecq, A.; Marrot, J.; Rivier
Keita, B.; Nadjo, L.; Secheresse, F. Inorg. Chem. 2007, 46 (13),
292−5301.
18) Mal, S. S.; Dickman, M. H.; Kortz, U.; Todea, A. M.; Merca, A.;
Bogge, H.; Glaser, T.; Muller, A.; Nellutla, S.; Kaur, N.; van Tol, J.;
Dalal, N. S.; Keita, B.; Nadjo, L. Chem.Eur. J. 2008, 14 (4),
186−1195.
19) Mitchell, S. G.; Gabb, D.; Ritchie, C.; Hazel, N.; Long, D. L.;
Cronin, L. CrystEngComm 2009, 11 (1), 36−39.
20) Bassil, B. S.; Ibrahim, M.; Mal, S. S.; Suchopar, A.; Biboum, R.
(
2
(
1
(43) Modec, B.; Brencic, J. V.; Zubieta, J. J. Chem. Soc., Dalton Trans.
2002, 1500−1507.
(44) Keita, B.; Lu, Y. W.; Nadjo, L.; Contant, R. Electrochem.
Commun. 2000, 2 (10), 720−726.
(
(
3
(
(
̀
e, E.;
́
5
(
̈
1
(
(
N.; Keita, B.; Nadjo, L.; Nellutla, S.; van Tol, J.; Dalal, N. S.; Kortz, U.
Inorg. Chem. 2010, 49 (11), 4949−4959.
(
21) Mitchell, S. G.; Boyd, T.; Haralampos, N. M.; De-Liang, L.;
Cronin, L. Inorg. Chem. 2011, 50, 136−143.
22) Chen, S.-W.; Boubekeur, K.; Gouzerh, P.; Proust, A. J. Mol.
Struct. 2011, 994, 104−108.
23) Zimmermann, M.; Belai, N.; Butcher, R. J.; Pope, M. T.;
Chubarova, E. V.; Dickman, M. H.; Kortz, U. Inorg. Chem. 2007, 46
5), 1737−1740.
24) Zhang, J.; Liu, T. B.; Mal, S. S.; Kortz, U. Eur. J. Inorg. Chem.
010, 20, 3195−3200.
25) Mal, S. S.; Nsouli, N. H.; Dickman, M. H.; Kortz, U. Dalton
Trans. 2007, 25, 2627−2630.
26) Muller, A.; Pope, M. T.; Todea, A. M.; Bogge, H.; van Slageren,
(
(
(
(
2
(
(
̈
J.; Dressel, M.; Gouzerh, P.; Thouvenot, R.; Tsukerblat, B.; Bell, A.
Angew. Chem., Int. Ed. 2007, 46 (24), 4477−4480.
(
27) Sousa, F. L.; Bogge, H.; Merca, A.; Gouzerh, P.; Thouvenot, R.;
Mu
28) Keita, B.; Kortz, U.; Holzle, L. R. B.; Brown, S.; Nadjo, L.
Langmuir 2007, 23 (19), 9531−9534.
̈
ller, A. Chem. Commun. 2009, 48, 7491−7493.
(
2
358
dx.doi.org/10.1021/ic202346r | Inorg. Chem. 2012, 51, 2349−2358