Chiral [Mo3S4H3(diphosphine) 3]+ hydrido clusters and study of the effect of the metal atom on the kinetics of the acid-assisted substitution of the coordinated hydride: Mo VS W

Article abstract of DOI:10.1021/ic100381u

The molybdenum(IV) cluster hydrides of formula [Mo₃S ₄H₃(diphosphine)₃]⁺ with diphosphine = 1,2-(bis)dimethylphosphinoethane (dmpe) or (+)-1,2-bis-(2R,5R)-2,5- (dimethylphospholan-1-yl)ethane ((R,R)-Me-BPE) have been isolated in moderate to high yields by reacting their halide precursors with borohydride. Complex [Mo₃S₄H₃((R,R)-Me-BPE)₃] ⁺ as well as its tungsten analogue are obtained in optically pure forms. Reaction of the incomplete cuboidal [M₃S₄H ₃((R,R)-Me-BPE)₃]⁺ (M = Mo, W) complex with acids in CH₂Cl₂ solution shows kinetic features similar to those observed for the related incomplete cuboidal [W₃S ₄H₃(dmpe)₃]⁺ cluster. However, there is a decrease in the value of the rate constants that is explained as a result of the higher steric effect of the diphosphine. The rate constants for the reaction of both clusters [M₃S₄H₃((R,R)-Me-BPE) ₃]⁺ (M = Mo, W) with HCl have similar values, thus indicating a negligible effect of the metal center on the kinetics of reaction of the hydrides coordinated to any of both transition metals.

Full text of DOI:10.1021/ic100381u

Inorg. Chem. 2010, 49, 5935–5942 5935
DOI: 10.1021/ic100381u
Chiral [Mo3S4H3(diphosphine)3]þ Hydrido Clusters and Study of the Effect of the
Metal Atom on the Kinetics of the Acid-Assisted Substitution of the Coordinated
Hydride: Mo vs W
†
†
‡
Andres G. Algarra, Manuel G. Basallote,*,† M. Jesus Fernandez-Trujillo, Marta Feliz,‡ Eva Guillamon,
ꢀ
ꢀ
ꢀ
ꢀ
Rosa Llusar,*,‡ Ivan Sorribes,‡ and Cristian Vicent§
†
´
´
ꢀ
Departamento de Ciencia de los Materiales e Ingenierıa Metaluꢀrgica y Quımica Inorganica, Facultad de
‡
ꢀ
ꢀ
´
´
Ciencias, Universidad de Cadiz, Apartado 40, Puerto Real, 11510 Cadiz, Spain, Departament de Quımica Fısica
§
´
ꢀ
i Analıtica, Universitat Jaume I, Av. Sos Baynat s/n, 12071 Castello de la Plana, Spain, and Serveis Centrals
ꢀ
ꢀ
d’Instrumentacio Cientifica, Universitat Jaume I, Av. Sos Baynat s/n, 12071 Castello de la Plana, Spain
Received February 25, 2010
The molybdenum(IV) cluster hydrides of formula [Mo3S4H3(diphosphine)3]þ with diphosphine = 1,2-(bis)dimethylphos-
phinoethane (dmpe) or (þ)-1,2-bis-(2R,5R)-2,5-(dimethylphospholan-1-yl)ethane ((R,R)-Me-BPE) have been isolated in
moderate to high yields by reacting their halide precursors with borohydride. Complex [Mo3S4H3((R,R)-Me-BPE)3]þ as well
as its tungsten analogue are obtained in optically pure forms. Reaction of the incomplete cuboidal [M3S4H3((R,R)-Me-
BPE)3]þ (M = Mo, W) complex with acids in CH2Cl2 solution shows kinetic features similar to those observed for the related
incomplete cuboidal [W3S4H3(dmpe)3]þ cluster. However, there is a decrease in the value of the rate constants that is
explained as a result of the higher steric effect of the diphosphine. The rate constants for the reaction of both clusters
[M3S4H3((R,R)-Me-BPE)3]þ (M = Mo, W) with HCl have similar values, thus indicating a negligible effect of the metal center
on the kinetics of reaction of the hydrides coordinated to any of both transition metals.
Introduction
of formula [W3Q4H3(diphosphine)3]þ with acids.7-9 These
studies include aspects such as solvent effects, ion pairing,
phosphine basicity, and chalcogen substitution. However, the
effect of the nature of the metal atom, Mo versus W, remains an
open question because the corresponding molybdenum hydride
complexes could not be isolated in pure form. The only evi-
dence of the existence of an incomplete cubane-type Mo3Q4
hydride cluster comes from Cotton’s group and goes back to
the late 1980s.10 The formulation of the [Mo3S4H3(diphos-
phine)3]þ complex obtained by reaction of its halide precursor
with borohydride is only based on spectroscopic evidence. In
this work, we report the isolation of two [Mo3S4H3(diphos-
phine)3]þ cluster salts, one of them in its optically pure form, a
circumstance that has allowed us to investigate the influence of
the group 6 metal atoms on the kinetics of acid-assisted
substitutions in these trinuclear cluster hydrides.
Molybdenum and tungsten are essential elements in biolo-
gical systems with oxidation numbers in the range from þ4
to þ6 in their reaction cycles.1 The aqueous chemistry of these
elements in their þ4 oxidation state is dominated by the
presence of trinuclear species, such as the incomplete cuboidal
M3O4 aquo ion, for which their sufides analogues exist.2,3
4þ
The motivation for studying these M3S4 complexes has come in
part from their structural relation to the mixed metal sulfido
clusters present in nitrogenase.4,5 Detailed studies by Sykes and
Hernandez-Molina on the kinetics and reaction mechanism of
the substitution of the terminal water ligands in the [M3Q4-
(H2O)9]4þ aquo ions (M=Mo, W and Q=S, Se) show a similar
behavior for Mo and W.6 In the recent past, our groups have
carried out a complete investigation on the kinetics and reac-
tion mechanism of the reaction of trinuclear cluster hydrides
The reaction of transition metal hydrides with acids
is a complex process that may involve different reaction
*To whom correspondence should be addressed. Tel.: þ34 964 728086 (R.L.).
Fax: þ34 964 728066 (R.L.). E-mail: manuel.basallote@uca.es (M.G.B.), Rosa.
Llusar@qfa.uji.es (R.L.).
(1) Sugimoto, H.; Tsukube, H. Chem. Soc. Rev. 2008, 37, 2609–2619.
(2) Bino, A.; Cotton, F. A.; Dori, Z. J. Am. Chem. Soc. 1978, 100, 5252–5253.
(3) Llusar, R.; Uriel, S. Eur. J. Inorg. Chem. 2003, 1271–1290.
(4) Georgiadis, M. M.; Komiya, H.; Chakrabarti, P.; Woo, D.; Kurnuc,
J. J.; Rees, D. C. Science 1992, 257, 1653–1659.
(7) Basallote, M. G.; Feliz, M.; Fernandez-Trujillo, M. J.; Llusar, R.;
Safont, V. S.; Uriel, S. Chem.;Eur. J. 2004, 10, 1463–1471.
(8) Basallote, M. G.; Estevan, F.; Feliz, M.; Fernandez-Trujillo, M. J.;
Hoyos, D. A.; Llusar, R.; Uriel, S.; Vicent, C. Dalton Trans. 2004, 530–536.
(9) Algarra, A. G.; Basallote, M. G.; Feliz, M.; Fernandez-Trujillo, M. J.;
Llusar, R.; Safont, V. S. Chem.;Eur. J. 2006, 12, 1413–1426.
(10) Cotton, F. A.; Kibala, P. A.; Matusz, M.; McCaleb, C. S.; Sandor,
R. B. W. Inorg. Chem. 1989, 28, 2623–2630.
(5) Kim, J.; Rees, D. C. Science 1992, 257, 1677–1682.
(6) Hernandez-Molina, R.; Sykes, A. G. J. Chem. Soc., Dalton Trans.
1999, 3137–3148.
r
2010 American Chemical Society
Published on Web 06/01/2010
pubs.acs.org/IC

5936 Inorganic Chemistry, Vol. 49, No. 13, 2010
Algarra et al.
intermediates because the actual reaction pathway depends
not only on the nature of the proton donor and acceptor but
also on the solvent and the presence and identity of the
counterion.11,12 In spite of the existing chemical analogies
between molybdenum and tungsten, differences arise regard-
ing bonding energies, acidity/basicity, and stability. An
increase in the homolytic M-H dissociation energies in
mononuclear cationic LnM-Hþ complexes when descending
within group 6 has been estimated.13 Relativistic effects are
considered to be the key factor responsible for the stronger
bond in the tungsten complex.
Lledos et al. have compared basicity, hydrogen bonding, and
the mechanism of reaction with acids for two homologues,
Mo(IV) and W(IV) metal hydrides of formula [Cp*M(dppe)-
H3].14 Both experimental and theoretical results show a greater
metal basicity for the tungsten system, although the two com-
pounds are reversibly protonated to yield an identical classical
tetrahydride product, [Cp*M(dppe)H4]þ, without the detection
of a dihydrogen intermediate. Contrary to the molybdenum sys-
tem, a direct proton transfer to the metal assisted by the hydride
ligand has been invoked for the reaction of [Cp*W(dppe)H3]
with acids. Regarding stability, the molybdenum hydride com-
plex loses H2 in coordinating solvents or in the presence of co-
ordinating anions, while its tungsten counterpart is stable under
the same conditions.11 Recent experiments by the same authors
indicate that the reaction of [Mo(CO)Cp*H(PMe3)2] with
methods.17,18 Diphosphine (R,R)-Me-BPE was purchased from
Strem Chemicals. Solvents for synthesis and electrochemical
measurements were dried and degassed by standard methods
before use.
Physical Measurements. Elemental analyses were performed
on an EA 1108 CHNS microanalyzer at the Universidad de La
Laguna. 31P{1H} NMR spectra were recorded on a Varian Mer-
cury 300 MHz apparatus and were referenced to external 85%
H3PO4. 1H, 13C{1H}, and 1H-13C gHSQC spectra were recorded
on a Varian INOVA 500 MHz apparatus using CD2Cl2 or acetone-
d6 as the solvent. Chemical shifts are reported in ppm from tetra-
methylsilane with the solvent resonance as the internal standard.
IR spectra were recorded on a Perkin-Elmer System 2000 FT-IR
using KBr pellets. Signal intensities are denoted as s=strong, m=
medium, and w=weak. Electronic absorption spectra were ob-
tained on a Perkin-Elmer Lambda-19 spectrophotometer in
dichloromethane. Circular dichroism measurements were recorded
on a JASCO J-810 spectropolarimeter. Electrospray mass spectra
were recorded with a Quattro LC (quadrupole-hexapole-quad-
rupole) mass spectrometer with an orthogonal Z-spray electro-
spray interface (Micromass, Manchester, U.K.). The cone voltage
was set at 20 V unless otherwise stated using CH3CN as the mobile
phase solvent. Nitrogen was employed as a drying and nebulizing
gas. Isotope experimental patterns were compared with theoretical
patterns obtained using the MassLynx 4.0 program.19
Synthesis. [Mo3S4H3(dmpe)3](BPh4), [1](BPh4). To a green
solution of [Mo3S4Cl3(dmpe)3](BPh4) (0.168 g, 0.130 mmol) in
THF (20 mL) was added an excess of NaBH4 (0.044 g, 1.164
mmol) under nitrogen. The solution color turned purple within
30 min. After the mixture was stirred for 2.5 h, it was taken to dry-
ness, redissolved in CH2Cl2, and filtered in order to eliminate the
excess of NaBH4 and other inorganic salts. Finally, a microcrystal-
line purple solid was obtained by slow diffusion of diethyl ether into
the filtrate (0.124 g, yield: 80%). Found Mo3S4P6C42H71B (%): S,
10.71; C, 42.26; H, 5.93. Requires (%): S, 10.79; C, 42.44; H,
6.02%. RMN 31P{1H}/δ: -144.00 (sept, 1J(P-F) 710.58 Hz),
24.00 (d, 2J(P-P) 15.9 Hz) and 41.12 (d, 2J(P-P) 15.9 Hz).
Et2O HBF4 leads to a dihydrogen complex in tetrahydrofurane
3
and to a classical dihydride in dichloromethane, thus showing
again the important role of the solvent in this kind of reaction.15
An essentially identical mechanism has been found for the
reaction between the Mo(IV) and W(IV) hydrides [Cp2MH2]
with acids.16 For both Mo and W complexes, acid attack occurs
in this case at the hydride site to generate dihydrogen species;
however, for the molybdenum product, the subsequent cleavage
of the dihydrogen ligand yields to the formation of dinuclear
species. Again, the higher instability of the Mo(IV) hydrides in
front of their tungsten homologues is evidenced.
2
RMN 1H/δ: -2.95 (3H hydride, dd, 2J(P-H) 62.8, J(P0-H)
36.5 Hz), 0.40 (9H, CH3, d, 2J(P-H) 8.5 Hz), 1.51 (9H, CH3, d,
2J(P-H) 8.5 Hz), 1.89 (3H, CH2, m,), 2.00 (9H, CH3, d, 2J(P-H)
8.5 Hz), 2.07 (3H, CH2, m), 2.19 (9H, CH3, d, 2J(P-H) 8.5 Hz),
2.51 (3H, CH2, m), 2.60 (3H, CH2, m,). RMN 13C{1H}/δ: 14.71
Despite the precedents above, systematic knowledge on the
kinetics of reaction with acids of transition metal hydrides of
the same stoichiometry for metals within the same group is still
scarce. With the isolation of the [Mo3S4H3(diphosphine)3]þ
cluster salts with two different diphosphines, dmpe or (R,R)-
Me-BPE, the limitation that precluded such systematic study
has been overcome. In this work, we report not only the
isolation of the first Mo3S4 cluster hydrides but also a
complete study on the kinetics of their reaction with acids
(HCl) in dichloromethane. The kinetics of the reaction are
compared with data obtained for their tungsten counterparts.
The study is also extended to the analysis of the phosphine
nature on the reaction kinetics and mechanism.
1
1
(CH3, d, J(C-P) 66.5 Hz), 20.66 (CH3, d, J(C-P) 136.5 Hz),
21.60 (CH3, d, 1J(C-P) 78.0 Hz), 21.82 (CH3, d, 1J(C-P) 62.0 Hz),
28.59 (CH2, m), 28.86 ppm (CH2, m). ESI-MS(CH3CN, 20 V): m/z
(%) 868.9 (100) [Mþ].
[Mo3S4H3((R,R)-Me-BPE)3]Cl ([2]Cl). To a suspension of
[Mo3S4Cl3((R,R)-Me-BPE)3]Cl (0,150 g, 0,112 mmol) in dry
THF (15 mL) was added LiBH4 (0.074 g, 3.398 mmol), and
then the mixture was stirred under a nitrogen atmosphere. After
48 h, the reaction mixture was filtered to remove the excess reducing
agent and inorganic salts formed. Then, solvent was removed under
a vacuum, and the solid was recrystallized from CH2Cl2/ether
mixtures. The resulting solid was washed with water, isopropanol,
and ether to yield 78 mg of the hydride product (yield: 56.4%).
Found Mo3ClS4P6C42H87 (%): C, 41,03; H, 7,15. Requires (%): C,
40,70; H, 6,91%. RMN 31P{1H} (CDCl3, 121 MHz) δ: 59.56 (d,
3P), 86.84 ppm (d, 3P). RMN 1H(CD2Cl2, 300MHz) δ:-2.37 (dd,
3H hydride, 2J(P-H) 36.44, 2J(P0-H) 63.37 Hz). RMN 13C{1H}
(CD2Cl2, 300 MHz) δ: 13.19 (CH3, d), 14.73 (CH3, d), 17.87 (CH3,
d), 18.04 (CH3, d), 24.98 (CH, d), 25.12 (CH, d), 26.03 (CH, d),
26.13 (CH, d), 35.12 (CH2, d), 35.75 (CH2, d), 38.37 (CH2, d), 39.0
Experimental Section
General Remarks. [Mo3S4Cl3(dmpe)3](BPh4) and [Mo3S4Cl3-
((R,R)-Me-BPE)3]Br were prepared according to literature
(11) Besora, M.; Lledos, A.; Maseras, F. Chem. Soc. Rev. 2009, 38, 957–966.
(12) Algarra, A. S. G.; Basallote, M. G.; Fernandez-Trujillo, M. J.;
Llusar, R.; Safont, V. S.; Vicent, C. Inorg. Chem. 2006, 45, 5774–5784.
(13) Wang, D. M.; Angelici, R. J. J. Am. Chem. Soc. 1996, 118, 935–942.
(14) Belkova, N. V.; Besora, M.;Baya, M.; Dub, P. A.; Epstein, L. M.; Lledos,
A.; Poli, R.; Revin, P. O.; Shubina, E. S. Chem.;Eur. J. 2008, 14, 9921–9934.
(15) Dub, P. A.; Belkova, N. V.; Filippov, O. A.; Daran, J.-C.; Epstein,
L. M.; Lledos, A.; Shubina, E. S.; Poli, R. Chem.;Eur. J. 2010, 16, 189–201.
(16) Henderson, R. A.; Oglieve, K. E. J. Chem. Soc., Dalton Trans. 1993,
3431–3439.
(CH2, d), 40.14 (CH2, d), 41.87 ppm (CH2, d). IR (KBr) cm-1
:
2923 (i), 2862 (i), 1450 (i), 1407 (m), 1369 (m), 1068 (m), 696 (m),
(17) Estevan, F.; Feliz, M.; Llusar, R.; Mata, J. A.; Uriel, S. Polyhedron
2001, 20, 527–535.
(18) Feliz, M.; Guillamon, E.; Llusar, R.; Vicent, C.; Stiriba, S. E.; Perez-
Prieto, J.; Barberis, M. Chem.;Eur. J. 2006, 12, 1486–1492.
(19) MASSLYNX, 4.0 ed.; Waters Ltd.: Milford, MA, 2005.

Article
Inorganic Chemistry, Vol. 49, No. 13, 2010 5937
Table 1. Crystallographic Data for [Mo3S4H3(dmpe)3](BPh4) 1/2CH2Cl2 ([1]BPh4 1/2CH2Cl2), [W3S4Br3((R,R)-Me-BPE)3]Br 1/2CH2Cl2 ([3]Br 1/2CH2Cl2), and
3
3
3
3
[W3S4H3((R,R)-Me-BPE)3]Br ([4]Br)
compound
empirical formula
fw
cryst syst
a, A
b, A
c, A
R, deg
β, deg
[1](BPh4) 1/2CH2Cl2
3
[3]Br 1/2CH2Cl2
3
[4]Br
C42.50H69BClMo3P6S4
1228.12
triclinic
C42.50H85Br4ClP6S4W3
1816.81
monoclinic
C42H87W3S4P6Br
1534.61
trigonal
15.2759(17)
15.9639(18)
16.1002(18)
119.268(2)
105.991(3)
98.261(3)
3105.8(6)
293(2)
P1
2
0.953
16886
1.47 to 25.00
12.0152(9)
19.7370(15)
15.0015(12)
14.9636(10)
23.254(3)
98.948(2)
γ, deg
V, A3
T, K
3520.7(5)
293(2)
P2(1)
2
7.476
19168
1.37 to 25.00
11606 [R(int) = 0.0544]
0.960
4509.2(7))
293(2)
R3
3
6.714
8163
1.80 and 26.00
3780 [R(int) = 0.0275]
1.056
space group
Z
μ(Mo KR), mm-1
reflns collected
φ range for data collection
unique reflns/Rint
Goodness-of-fit on F2
R1a/wR2b
R1a/wR2b (all data)
residual F/e A-3
10441 [R(int) = 0.0338]
1.125
R1 = 0.0745, wR2 = 0.2252
R1 = 0.1010, wR2 = 0.2459
3.063 (0.99 A from Mo) and -0.662
R1 = 0.0564, wR2 = 0.1484
R1 = 0.0815, wR2 = 0.1684
3.943 (0.96 A from W) and -1.502
R1 = 0.0301, wR2 = 0.0820
R1 =0.0350, wR2 = 0.0863
1.646 and -1.041
P
P
P
P
a R1 = ||Fo| - |Fc||/ Fo. b wR2 = [ [w(F2o - F2c)2]/ [w(Fo2)2]]1/2
.
635 (m), 458 (d). CD (1.15 ꢀ 10-4 M, CH2Cl2) λ nm (mdeg): 531
(-36), 347 (-9), 288 (53), 240 (-74), 231 (43). UV-vis (CH2Cl2):λ
(ε) 548 (1414), 436 (3545), 387 (3975.41), 244 nm (40 540 mol-1 m3
cm-1). ESI-MS (CH3CN, 20 V): m/z (%) 1193 (100) [Mþ].
[W3S4Br3((R,R)-Me-BPE)3]Br ([3]Br). This compound was
prepared by an excision reaction of the polymer {W3S7Br4}n (0.200
g, 0.183 mmol with (R,R)-Me-BPE (0.212 g, 0.821 mmol)) in 20
mL of dried CH3CN under reflux. After 24 h, the reaction mixture
was filtered, the filtrated was taken to dryness under a vacuum, and
the solid was dissolved in dichloromethane. The addition of ether
(30 mL) caused the complete precipitation of a blue solid, which
was recrystallized from CH2Cl2/ether mixtures and washed with
toluene/acetone (95:5) mixtures to afford 307 mg of the desired
product (yield: 95%). Found W3Br4S4P6C18H48 (%): S, 7.22; C,
28.43; H, 4.78. Requires (%): S, 6.93; C, 28.37; H, 4.72. RMN
31P{1H} (CD2Cl2, 121 MHz) δ: 31.69 (dd, 3P, 2JP-Pgem = 2.56 Hz,
W-H), 953 (i), 940 (i), 848 (i), 557 (d), 439 (d, W-μ3S), 423 (d,
W-μ3S), 355 (d). CD (1.18 ꢀ 10-4 M, CH2Cl2) λ nm (mdeg): 482
(-52), 394 (28), 308 (-45), 269 (þ11), 235 nm (-47). UV-vis
(CH2Cl2): λ (ε) 502 (1654), 311 (8426), 236 nm (27891 mol-1 m3
cm-1). ESI-MS (CH3CN, 10 V) m/z (%): 1457 (100) [Mþ].
Kinetic Experiments. The kinetics of reaction of the 2þ and 4þ
clusters with acids were studied at 25.0 °C using an Applied Photo-
physics SX17MV stopped-flow instrument provided with a PDA1
diode-array detector, and the results were analyzed with the
SPECFIT program.20 All experiments were carried out in dichlor-
omethane solutions of the cluster salt and the acid. The complex
solutions were prepared at concentrations (0.8-9.0)ꢀ10-4 mol
3
dm-3, and preliminary experiments at two or three different con-
centrations were carried out in all cases to confirm the first order
dependence of the observed rate constants with respect to the
complex concentration. Solutions of HCl in dichloromethane were
prepared by mixing the required amounts of chlorotrimethylsilane
and methanol. The acid solutions were used within 2-3 h from
preparation, and their concentrations were determined by titration
with KOH (phenolphthalein indicator) of solutions resulting from
adding an aliquot (3 mL) to 50 mL of water and stirring vigorously
for 20 min. Most experiments were carried out under pseudo-first-
order conditions of acid excess, and the analysis required the use of
a multistep kinetic model.
3JP-Ptrans = 3.70 Hz), 35.90 (dd, 3P, 2JP-Pgem =2.64 Hz, 3JP-Ptrans
=
3.45 Hz) (AA0A00BB0B00 system). RMN 13C{1H} (CD2Cl2, 300
MHz) δ: 15.41 (CH3, d), 16.85 (CH3, d), 19.12 (CH3, d), 20.79
(CH3, d), 25.47 (CH, d), 25.61 (CH, d), 27.81 (CH, d), 27.99 (CH,
d), 32.60 (CH2, d), 35.5 (CH2, s), 35.8 (CH2, d), 39.75 (CH2, d),
41.96 (CH2, d), 43.10 ppm (CH2, d). IR (KBr) cm-1: 2923 (i), 2858
(i), 2130 (d), 1453 (i), 1410 (m), 1066 (m), 924 (i), 730 (d), 630 (m),
453(m). CD(1.13ꢀ 10-4 M, CH2Cl2) λnm (mdeg): 543 (-33), 405
(-62), 326 (89), 292 (-28), 271 (143), 248 (175). UV-vis
(CH2Cl2): λ (ε) 325 (13 360), 278 (20 190), 238 nm (31 580 mol-1
m3 cm-1). ESI-MS (CH2Cl2, 20 V): m/z(%) = 1694 (100) [Mþ].
[W3S4H3((R,R)-Me-BPE)3]Br ([4]Br). This compound was
prepared following the same procedure for [2]Cl but in this case
using P-[W3S4Br3((R,R)-Me-BPE)3]Br (0.060 g, 0.034 mmol) as a
starting material and LiBH4 (0.022 g, 1.010 mmol) in 10 mL of dry
THF. The reaction occurred with a color change from blue to pink,
and after 6 h the mixture was filtered. The filtrate was taken to
dryness and then was dissolved in a minimum of dichloromethane.
The addition of ether caused the complete precipitation of the pink
product, which was then washed with water, isopropanol, and ether
successively to afford 39 mg of P-[W3S4H3((R,R)-Me-BPE)3]Br
(yield: 75%). Found W3S4P6C48H87Br (%): S, 8.32; C, 32.80; H,
5.71. Requires (%): S, 7.82; C, 32.45; H, 5.63. RMN 31P{1H} (CD2-
Cl2, 121 MHz) δ: 27.04 (s, 3P), 56.58 (s, 3P). RMN 1H (CD2Cl2, 300
MHz) δ: -0.408 (dd, 3H hydride, 2J(P-H) 29.91, 2J(P0-H) 47.65
Hz). RMN 13C{1H} (CD2Cl2, 300 MHz) δ: 12.48 (CH3, d), 14.01
(CH3, d), 16.99 (CH3, d), 18.66 (CH3, d), 25.88 (2 CH, c), 27.03 (2
CH, c), 34.79 (CH2, d), 35.05 (CH2, s), 37.91 (CH2, d), 39.61 (CH2,
d), 40.25 (CH2, d), 43.17 ppm (CH2, d). IR (KBr) cm-1: 1709 (m,
X-Ray Studies. The crystals are air stable and were mounted
on the tip of a glass fiber with the use of epoxy cement. X-ray
diffraction experiments were carried out on a Bruker SMART
CCD diffractometer using Mo KR radiation (λ = 0.71073 A) at
room temperature. The data were collected with a frame width
of 0.3° in ω and a counting time of 10 s per frame. The diffraction
frames were integrated using the SAINT package and corrected
for absorption with SADABS.21,22 The structures were solved
by direct methods and refined by the full-matrix method based
on F2 using the SHELXTL software package.23 The crystal
parameters and basic information relating data collection and
structure refinement for compounds [1](BPh4) 1/2CH2Cl2,
3
[3]Br 1/2CH2Cl2 and [4]Br are summarized in Table 1.
3
€
(20) Binstead, R. A.; Jung, B.; Zuberbuhler, A. D. SPECFIT; Spectrum
Software Associates: Chappel Hill, NC, 2000.
(21) SAINT, version 6.2; Bruker Analytical X-Ray Systems: Madison, WI, 2001.
€
€
(22) Sheldrick, G. M. SADABS; University of Gottingen: Gottingen,
Germany, 1996.
(23) Sheldrick, G. M. SHELXTL, version 5.1; Bruker Analytical X-Ray
Systems: Madison, WI, 1997.

5938 Inorganic Chemistry, Vol. 49, No. 13, 2010
Algarra et al.
For compound [1](BPh4) 1/2CH2Cl2, all atoms in the 1þ cluster
3
cation and the BPh4- anion were refined anisotropically. The posi-
tions of all hydrogen atoms in the diphosphine ligands and BPh4
-
anion were generated geometrically, assigned isotropic thermal
parameters, and allowed to ride on their respective parent carbon
atoms. In the last stages of the refinement, half a molecule of
CH2Cl2 was located and refined isotropically as a rigid group.
Hydrogen atoms from this CH2Cl2 molecule were fixed at calcu-
lated positions and included in the refinement. For compound
[3]Br 1/2CH2Cl2, all atoms in the 3þ cluster cation and the
3
bromine anion were refined anisotropically. Several C-C bond
distances of the noncrystallographically equivalent five-membered
rings of the Me-BPE ligands (C(7)-C(8), C(9)-C(10), C(11)-C-
(12), C(13)-C(14), C(18)-C(19), C(24)-C(25), C(26)-C(27),
and C(29)-C(30)) were constrained to a fixed value (typically
1.48 A). The positions of all hydrogen atoms in the diphosphine
ligands were generated geometrically, assigned isotropic thermal
parameters, and allowed to ride on their respective parent carbon
atoms. In the last stages of the refinement, half a molecule of
CH2Cl2 was located and refined isotropically as a rigid group.
Hydrogen atoms from this CH2Cl2 molecule were fixed at calcu-
lated positions and included in the refinement. For compound
[4]Br, all atoms in the 4þ cluster cation and the bromine anion were
refined anisotropically. The C(3)-C(4) bond distance of one five-
membered ring of the Me-BPE ligands was constrained to a fixed
value (typically 1.48 A). The positions of all hydrogen atoms in the
diphosphine ligands were generated geometrically, assigned iso-
tropic thermal parameters, and allowed to ride on their respective
parent carbon atoms. A high electronic positive residual density ca.
3.0 e/A3 for [1]BPh4 and ca. 3.9 e/A3 for [3]Br stage remained in the
last refinement, and in both cases these residual peaks were located
at distances below 1 A from the Mo(3) atom and W(1), respec-
tively. A common feature of all three crystal structures is a
moderate to high difference of the thermal parameters between
neighbor atoms in the diphosphane backbone, dmpe for 1þ and
Me-BPE for 3þ and 4þ. Such differences are attributed to the
commonly found solid state disorder in diphosphane-ethane
complexes, which exists with two limit configurations: one with
the two diphosphine backbone carbon located on different sides of
the MoP(1)P(2) plane and the other with one of these two carbon
atoms situated essentially in the MoP(1)P(2) plane.24
Figure 1. ORTEP representation of the cationic cluster [Mo3S4H3-
(dmpe)3]þ (1þ).
Table 2. Selected Averaged Bond Distances (A) for Compounds [Mo3S4H3-
(dmpe)3](BPh4) 1/2CH2Cl2 ([1]BPh4 1/2CH2Cl2), [W3S4Br3((R,R)-Me-BPE)3]Br 1/
3
3
3
3
2CH2Cl2 ([3]Br 1/2CH2Cl2), and [W3S4H3((R,R)-Me-BPE)3]Br ([4]Br)
[Mo3S4Cl3
(dmpe)3(PF6)b
dist. (A)a
[1](BPh4)
[3]Br
[4]Br
M-M
2.744[5]
2.346[4]
2.329[4]
2.324[2]
2.467[12]
2.524[9]
2.766(4)
2.360(9)
2.336(7)
2.290(7)
2.534(8)
2.605(8)
2.793[4]
2.371[4]
2.289[5]
2.311[6]
2.559[3]
2.633[7]
2.7753(5)
2.364(3)
2.350(2)
2.334(2)
2.491(2)
2.548(2)
M-(μ3-S)
M-(μ-S)c
M-(μ-S)d
M-P(1)e
M-P(2) f
a Standard deviations for averaged values are given in brackets.
b Data taken from ref 26. c Distance trans to the Mo-P bond. d Distance
trans to the Mo-X bond. e Distance trans to the M-(μ3-S) bond.
f Distance trans to the M-(μ-S) bond.
with a 9-fold excess of NaBH4 in THF. An ORTEP drawing
of the 1þ cation is represented in Figure 1.
Results and Discussion
The molybdenum and sulfur atoms in 1þ occupy adjacent
vertices in a cube with a metal position missing, which
results in an incomplete cubane-type structure. The three
metal atoms define an approximately equilateral triangle
with Mo-Mo bond distances of 2.744 [5] A in agreement
with the presence of a single metal-metal bond and a þ4
oxidation state for the metal. The nature of the Mo3(μ3-S)-
(μ-S)3 core in 1þ is such that the bridging and capping
sulfur atoms occupy a set of facial positions around the
octahedrally coordinated metal atoms, leaving the three
outer facial sites available for the two phosphorous atoms
of the diphosphine and an apparently empty site occupied
by a hydrogen atom. The existence of three hydride ligands,
Synthesis, Structure, and Reactivity. Molybdenum
hydride complexes are proved more difficult to isolate than
their tungsten congeners as a result of their inherent instabi-
lity.25 The first reported synthesis of a cuboidal W3S4 tri-
nuclear cluster hydride appeared in 1989 by conversion of
the chloro [W3S4Cl3(diphoshine)3]þ (diphosphine = dmpe,
depe) species to hydrido species according to eq 1.26
½W3S4Cl3ðdiphosphineÞ ꢁþ þ BH4 ðexcessÞ
-
3
f ½W3S4H3ðdiphosphineÞ ꢁþ
ð1Þ
3
Attempts to extend this procedure to the analogous
molybdenum system have not provided a definite proof of
the existence of these molybdenum hydrides, although spec-
troscopic evidence of the formation of [Mo3S4H3(dmpe)3]þ
(1þ) cations using 31P{1H} NMR has been pointed out.10
We have now been able to isolate the [Mo3S4H3(dmpe)3]-
(BPh4) salt in 80% yields by reacting its chloro precursor
1
one on each Mo atom, has been fully supported by H
NMR spectroscopy. The proton spectrum of [Mo3S4H3-
(dmpe)3]þ shows a doublet of doublets in the hydride region
centered at δ = -2.95 ppm, attributed to the splitting of the
two nonequivalent phosphorous atoms bonded to the same
metal with 2J(P-H) coupling constants of 62.8 and 36.5 Hz.
Table 2 contains a list of important bond distances and
angles. The metal-metal and metal-sulfur distances within
the Mo3S4 cluster core follow the tendencies observed for
other trinuclear M3Q4 (M = Mo, W; Q = S, Se) species.
The Mo-(μ3-S) distance in 1þ is approximately 0.02 A
(24) Zhang, Y. P.; Bashkin, J. K.; Holm, R. H. Inorg. Chem. 1987, 26,
694–702.
(25) Minato, M.; Ito, T. Coord. Chem. Rev. 2008, 252, 1613–1629.
(26) Cotton, F. A.; Llusar, R.; Eagle, C. T. J. Am. Chem. Soc. 1989, 111,
4332–4338.

Article
Inorganic Chemistry, Vol. 49, No. 13, 2010 5939
Figure 3. Circular dichroism spectra of P-[W3S4Br3((R,R)-Me-BPE)3]-
Br ([3]Br; ---) and P-[W3S4H3((R,R)-Me-BPE)3]Br ([4]Br, ;) in dichloro-
methane at 25 °C.
Figure 2. Circular dichroism spectra of P-[Mo3S4Cl3((R,R)-Me-BPE)3]-
Cl (---) and P-[Mo3S4H3((R,R)-Me-BPE)3]Cl (P-[2]Cl; ;) in dichloro-
methane at 25 °C.
system, indicating that stereochemistry is preserved
during the reaction. ORTEP views of 3þ and 4þ cluster
P enantiomers, where the capping sulfur atoms point
against the viewer, are represented in Figure 4. A sum-
mary of selected bond distances for [3]Br and [4]Br is
given in Table 2. Again, we observe that substitution of a
bromide ligand in 3þ by hydrogen produces an increase in
the Mo-(μ-S) distance trans to that position of approxi-
mately 0.023 A due to the higher trans influence of the
hydride versus the bromide ligand. Complexes 3þ and 4þ
share structural features with other [M3Q4X3(diphos-
phine)3]þ compounds reported up to date.3 The me-
tal-metal distances are in the range (2.73-2.83 A) found
for other M3S4 complexes. The two types of W-P dis-
tances differ by 0.06-0.07 A with the one trans to the
capping sulfur atom being shorter.
longer than the average Mo-(μ-S) bond lengths, and there
are two kinds of Mo-(μ-S) distances. The substitution
of a chloride ligand in the [Mo3S4Cl3(dmpe)3]þ starting
material by hydrogen is reflected in the Mo-(μ-S) distance
trans to that position that increases by approximately 0.034
A in the hydrido cluster due to the higher trans influence of
the hydride versus the chloride ligand. The specific coordi-
nation of the diphosphine ligand in 1þ, with one phos-
phorous atom trans to the capping sufur atom and the other
trans to the bridging sulfur atom, results in cubane-type
sulfido clusters with backbone chirality. In this case, com-
plex 1þ is obtained as a racemic mixture of the P and M
enantiomers. The P and M symbols refer to the rotation of
the H atom around the C3 axis, with the capping sulfur atom
pointing toward the viewer.
Following the same synthetic strategy used for the
preparation of 1þ, we have also been able to isolate
the optically pure P-[Mo3S4H3((R,R)-Me-BPE))3]Cl salt
([2]Cl) also starting from its chloride precursor, this time
by reaction with lithium borohydride, in 56% yield.
Furthermore, the stereochemistry of the starting material
is preserved in the final product as evidenced from the
identical signal pattern of the circular dichroism spectra,
represented in Figure 2. The circular dichroism spec-
trum of [2]Cl shows two signals at λmax=267 and 417 nm
for þ101 and þ35 mdeg, respectively, as in the case of
its P-[Mo3S4Cl3((R,R)-Me-BPE))3]þ precursor that also
shows two bands at λmax = 265 and 416 nm for þ290
and þ35 mdeg, respectively.
The availability of a series of complexes [M3S4H3-
(diphosphine)3]þ with M = Mo and W containing two
different diphosphine ligands, dmpe and (R,R)-Me-BPE,
provides a unique opportunity to systematically investi-
gate the influence of the metal and diphosphine basicity
on the kinetics and reaction mechanisms of these hydrido
species with acids. The kinetics of reaction with acids of
[W3Q4H3(dmpe)3]þ (Q = S, Se) cations in coordinating
and noncoodinating solvents have been studied pre-
viously. In particular, when the reaction is carried out
in dichloromethane, two competitive pathways showing a
first and a second order dependence with respect to the
acid are found, the second order pathway being the
preferred one under pseudo-first-order conditions of acid
excess.7-9,12 Unfortunately, comparison with the Mo
analogue was not possible because the reaction of the
[Mo3S4H3(dmpe)3]þ (1þ) hydride with acids could not be
followed by stopped-flow techniques even under rigorous
air-free conditions. No reproducible kinetic traces could
be obtained for this reaction, most likely due to the lack of
stability of the intermediates toward traces of oxygen.
Fortunately, the molybdenum cluster 2þ containing the
chiral diphosphine (R,R)-Me-BPE was more stable and
allowed kinetic studies.
For comparative purposes, the tungsten P-[W3S4H3-
((R,R)-Me-BPE)3]Br ([4]Br) analogue was synthesized
starting from its bromide precursor, which in turn was
prepared by excision of the polymeric {W3S7Br4}n phases
in acetonitrile in the presence of the optically pure (R,R)-
Me-BPE diphosphine. As reported for the molybdenum
system, the reaction turned out to be enantioselective with
the exclusive formation of the P-[W3S4Br3((R,R)-Me-
BPE)3]Br ([3]Br) enantiomer as judged by X-ray single
crystal analysis and circular dichroism spectroscopy.18
The [4]Br tungsten hydride presents the same signal se-
quence as its halide [3]Br precursor in its circular dichro-
ism spectrum (see Figure 3), as seen in the molybdenum
The molybdenum and tungsten trinuclear clusters
[Mo3S4H3((R,R)-Me-BPE)3]þ (2þ) and [W3S4H3((R,R)-
Me-BPE)3]þ (4þ) react with HCl in CH2Cl2 to form

5940 Inorganic Chemistry, Vol. 49, No. 13, 2010
Algarra et al.
Figure 4. ORTEP representations of the cationic clusters P-[W3S4Br3((R,R)-Me-BPE)3]þ (3þ; left) and P-[W3S4H3((R,R)-Me-BPE)3]þ (4þ; right).
Figure 5. Typical spectral changes with time for the reaction of clusters 2þ (left) and 4þ (right) with HCl in CH2Cl2 solution at 25.0 °C.
[M3S4Cl3((R,R)-Me-BPE)3]þ, as represented in eq 2.
been discussed recently for the related [W3PdS4H3(dmpe)3-
(CO)]þ.27,28
The Reaction of Clusters 2þ and 4þ with Acids: Kinetics
of Reaction with HCl. In order to analyze the influence of
the metal on the acid-assisted substitution process repre-
sented in eq 2, the kinetics of the reactions between clus-
ters 2þ and 4þ with an excess of HCl in CH2Cl2 were inves-
tigated, and typical results of the stopped-flow experiments
using a diode-array detector are illustrated in Figure 5. The
formation of P-[M3S4Cl3((R,R)-Me-BPE)3]þ (M=Mo, W)
as the final reaction products was confirmed in indepen-
dent NMR experiments. A detailed analysis of the spectral
changes of the type shown in Figure 5 reveals that the
reaction of both clusters with HCl occurs with three kineti-
P-½M3S4H3ððR, RÞ-Me-BPEÞ ꢁþ þ 3HCl
3
f P-½M3S4Cl3ððR, RÞ-Me-BPEÞ ꢁþ þ 3H2
ð2Þ
3
No reaction is observed between these molybdenum and
tungsten cluster hydrides and halide salts, as previously
found for the [W3Q4H3(dmpe)3]þ (Q=S, Se) system, sug-
gesting that dihydrogen-bonded species are required to be
formed as reaction intermediates in the course of the reac-
tion. In contrast with the dmpe derivatives, the Mo and W
hydride complexes containing the (R,R)-Me-BPE dipho-
sphine ligand do not react with HCl in acetonitrile or
acetonitrile/water mixtures, which shows that there is a
significant basicity decrease in the hydride ligands when
replacing dmpe for the less basic (R,R)-Me-BPE ligand.
Fortunately, both compounds react with excess HCl in
dichloromethane solutions, and this allowed a kinetic study
on the effect of the nature of the metal center. Such a
dramatic effect of solvent on the reactions of this kind of
hydride clusters is not unprecedented, and the reasons that
lead to the lack of reaction in acetonitrile solutions have
cally distinguishable steps with rate constants k1obs, k2obs
,
and k3obs. For a given concentration of acid, no changes in
the observed rate constants were observed when the con-
centration of any of both clusters was changed, which
(27) Basallote, M. G.; Algarra, A. G.; Feliz, M.; Fernandez-Trujillo,
M. J.; Llusar, R.; Safont, V. S. Chem.;Eur. J. 2010, 1613–1623.
(28) Algarra, A. G.; Basallote, M. G.; Feliz, M.; Fernandez-Trujillo,
M. J.; Guillamon, E.; Llusar, R.; Vicent, C. Inorg. Chem. 2006, 45, 5576–
5584.

Article
Inorganic Chemistry, Vol. 49, No. 13, 2010 5941
the electronic spectra of the intervening species in the case of
[W3S4H3(dmpe)3]þ and 4þ indicate that the mechanism of
reaction between the acid and the coordinated hydride in 4þ
is the same as that for the previously studied [W3S4H3-
(dmpe)3]þ cluster, with sequential participation of two HCl
molecules in the initial formation of the dihydrogen-
bonded species (eqs 4 and 5).9 The second HCl molecule
forms a hydrogen bond with the first one, which causes a
decrease of the H H distance that facilitates H2 release in
3 3 3
the rate-determining step (eq 6).9,12,27 If the initial equilibria
are displaced toward the starting reagent, the rate law for
the processes in eqs 4-6 has the same form as that of eq 3
with the equivalence ki=Ke1Ke2k3. As the process would be
repeated for the three metal centers, three steps with a
second order dependence with respect to the acid will be
observed. It is interesting to note that significant deviations
from statistically controlled kinetics are also observed; i.e.,
the rates of reaction at the three metal centers are in a ca.
1000:40:1 ratio, very far from the 3:2:1 ratio expected from
statistical considerations. Although Sykes and Hernandez-
Molina reported many examples that established firmly
that substitution reactions in M3S4 aqua cluster complexes
occur with statistically controlled kinetics,6 we have found
that deviations appear in some cases, especially when
reactions are carried out in organic solvents. On the other
hand, the values of k1 in Table 3 for the W complexes 4þ
and [W3S4H3(dmpe)3]þ indicate that substitution of dmpe
by a bulkier diphosphine such as (R,R)-Me-BPE leads to a
decrease in the rate of reaction of about 1 order of
magnitude. With regard to the effect of the metal center,
the values in Table 3 indicate that both clusters 2þ and 4þ
react with similar rates with HCl, the quotient kW/kMo
between rate constants corresponding to the same resolved
kinetic step, being in all cases close to 2.
Figure 6. Electronic spectra calculated for the starting complex, the
reaction intermediates, and the final reaction product from the spectral
changes observed during the reaction of cluster 2þ (left) and 4þ (right)
with HCl in CH2Cl2 solution at 25.0 °C.
Figure 7. Plots of the observed rate constants versus [Hþ] for the three
stepsinthe reaction of2þ (left) and3þ (right) withHClinCH2Cl2 solution
at 25 °C.
Table 3. Rate Constants of the Different Steps Observed during the Protonation
of 2þ and 4þ with HCl in CH2Cl2 at 25.0 °Ca
2þ
4þ
[W3S4H3(dmpe)3]þb
10-4 ꢀ k1 (M-2 s-1
10-2 ꢀ k2 (M-2 s-1
)
)
1.28 ( 0.07 1.9 ( 0.1
24.1 ( 0.6
103 ( 3
c
4.3 ( 0.2
11.2 ( 0.5
29.8
10 ( 3
23 ( 2
19
k3 (M-2 s-1
k1/k2
k1/k3
)
2
23.4
kiobs ¼ ki½HXꢁ
ð3Þ
ð4Þ
1143
38.4
826
43.5
k2/k3
a For comparative purposes information about the cluster
[W3S4H3(dmpe)3]þ has also been included. b Data taken from ref 9.
c The third step of this reaction is independent of the HCl concentration
M- H þ HCl h M- H H- Cl; Ke1
3 3 3
and has a value of 4 ( 1 10-3 s-1
.
M- H H- Cl þ HCl h M- H H- Cl H- Cl; Ke2
3 3 3 3 3 3 3 3 3
indicates that all three steps show a first-order depen-
dence on the cluster concentration. The analysis of the
kinetic data also provides the electronic spectra calculated
for the starting complexes (2þ or 4þ), the reaction inter-
mediates, and the final reaction products, which are shown
in Figure 6. It is important to highlight, that in the case of the
tungsten complex, all of the calculated spectra are very
similar to those obtained for the reaction of the related
[W3S4H3(dmpe)3]þ cluster under the same conditions.9
For both complexes, the rate constants derived for the
three steps change with the acid concentration according to
eq 3 (see Figure 7), and the values of the corresponding third
order rate constants as well as the k1/k2, k1/k3, and k2/k3
ratios are shown in Table 3. Attempts to improve the quality
of the fits by including additional zero and first order terms
with respect to the acid were unsuccessful because they led
inevitably to negative values of those terms. On the other
hand, alternative dependencies such as [Hþ][Cl-] can be
ruled out9 because HCl is a weak acid in organic solvents
such as dichloromethane. The second order dependence with
respect to the HCl concentration and the similarities between
ð5Þ
M- H H- Cl H- Cl f M- Cl þ H2 þ HCl; k3
3 3 3
3 3 3
ð6Þ
Conclusions
In this paper, we report the synthesis of a novel hydride
cluster of formula [M3S4H3((R,R)-Me-BPE)3]þ (M=Mo, W;
(R,R)-Me-BPE = (þ)-1,2-bis-(2R, 5R)-2,5-(dimethylphos-
pholan-1-yl)ethane). The molybdenum(IV) cluster hydride of
formula [Mo3S4H3(dmpe)]þ (dmpe=(1,2-(bis)dimethylphos-
phinoethane) has also been isolated. Both [M3S4H3((R,R)-
Me-BPE)3]þ (M = Mo, W) compounds react with HCl in
CH2Cl2 in a way similar to that previously observed for
[W3S4H3(dmpe)]þ, three kinetically resolved steps being ob-
served under acid excess and corresponding to the sequential
substitution of the three hydride ligands by the anion of the acid.
The values of the rate constants indicate that the steric effect
associated with the bulkier (R,R)-Me-BPE chiral diphosphine

5942 Inorganic Chemistry, Vol. 49, No. 13, 2010
Algarra et al.
ꢀ
leads to a decrease in the rate constants as well as higher devia-
tions of the kinetics with respect to the statistical behavior. The
rate constants of the reaction between both [M3S4H3((R,R)-Me-
BPE)3]þ (M = Mo, W) compounds with HCl in CH2Cl2 are
very similar, thus showing that the nature of the transition metal
does not modify to a great extent the kinetics of the process.
However, as the observed rate constants contain contributions
from the initial formation of dihydrogen bonded adducts (eqs 4
and 5) and the rate-determining elimination of H2 (eq 6), it is not
possible to decide if the close values of the rate constants found for
both metals indicate that there are no significant changes in any of
the mechanistic steps or they result from compensating changes
in the constants corresponding to the different contributions.
and CTQ2006-14909-C02-01), Fundacio Bancaixa-UJI
(research project P1.1B2007-12), Generalitat Valenciana
(ACOMP/2009/105 and Prometeo/2009/053), and Junta de
´
Andalucıa (Grupo FQM-137, Grant P07-FQM-02734) is
gratefully acknowledged. The authors also thank the Servei
ꢀ
Central D’Instrumentacio Cientifica (SCIC) of the Universi-
tat Jaume I and the Servicios Centrales de Ciencia y Tecno-
ꢀ
logıa of the Universidad de Cadiz for providing us with the
´
mass spectrometry, NMR, and X-ray facilities. I.S. thanks
ꢀ
the Spanish Ministerio de Ciencia e Innovacion (MICINN)
for a doctoral fellowship (FPU).
Supporting Information Available: Crystallographic data
(excluding structure factors) for the structures reported in this
paper. This material is available free of charge via the Internet at
http://pubs.acs.org.
Acknowledgment. The financial support of the Spanish
ꢀ
Ministerio de Ciencia e Innovacion (Grants CTQ2008-02670