Preparation and properties of the heterometallic cube [Mo3CdS4(H2O)12]4+ and the reaction with H+

Article abstract of DOI:10.1039/b109968j

The orange-brown cadmium-containing heterometallic cube [Mo₃CdS₄(H₂O)₁₂]⁴⁺ has been prepared by reacting [Mo₃S₄(H₂O)₉]⁴⁺ in 2.0 M HCl with Cd

Full text of DOI:10.1039/b109968j

View Article Online / Journal Homepage / Table of Contents for this issue
Preparation and properties of the heterometallic cube
[Mo3CdS4(H2O)12]4؉ and the reaction with H؉
Iain J. McLean, Maxim N. Sokolov, Rita Hernandez-Molina and A. Geoffrey Sykes*
Department of Chemistry, The University of Newcastle, Newcastle upon Tyne, UK NE1 7RU
Received 31st October 2001, Accepted 24th January 2002
First published as an Advance Article on the web 3rd April 2002
The orange–brown cadmium-containing heterometallic cube [Mo3CdS4(H2O)12]4ϩ has been prepared by reacting
[Mo3S4(H2O)9]4ϩ in 2.0 M HCl with Cd2ϩ in the presence of H3PO2 as reductant for ∼1 h at 20 ЊC. Alternatively, it can
be obtained by heating a solution of [Mo3S4(H2O)9]4ϩ in 0.5 M Hpts (ptsϪ = p-toluenesulfonate) with Cd metal for
∼1 h at 70 ЊC. The cube reacts with Hϩ in a process which is inhibited by the replacement of ClϪ with ptsϪ. Purification
by elution from a Dowex cation-exchange column was carried out at low [Hϩ] using a mixture of 2.9 M Lipts and 0.1 M
Hpts. On titration with [Co(dipic)2]Ϫ, the incomplete cube [Mo3S4(H2O)9]4ϩ is reformed, and two equivalents of CoIII
oxidant are consumed per cube, consistent with the equation Mo3CdS44ϩ ϩ 2CoIII
Mo3S44ϩ ϩ Cd2ϩ ϩ 2CoII.
Reaction with [Hϩ] in the range 0.5–1.5 M, I = 2.0 M (Lipts) gives H2, which was determined quantitatively by
gas chromatography. The kinetics of the decay Mo3CdS44ϩ ϩ 2Hϩ
Mo3S44ϩ ϩ Cd2ϩ ϩ H2 give a rate law
kH[Mo3CdS44ϩ][Hϩ]2, with kH = 4.6 × 10Ϫ4 MϪ2 sϪ1 at 25 ЊC, I = 2.00 M (Lipts). In 1 M HCl, I = 2.0 M (Lipts),
the half-life of 0.3 mM solutions is ∼5 min.
603 (360) in 2 M Hpts.10 Purification and concentration of the
product were carried out by Dowex 50W-X2 cation-exchange
chromatography using standard procedures.
Introduction
The aim of this paper is to report procedures for the formation
of a heterometallic derivative by incorporating Cd into the
incomplete (metal-depleted) [Mo3S4(H2O)9]4ϩ cube,1 and com-
ment on the unusual sensitivity of the cube towards reaction
with Hϩ. Of the group 12 elements, Hg has been studied most
and provides the only example to date of a fully characterised
Other materials
Hypophosphorous acid, H3PO2 (50% w/w H2O solution), p-
toluenesulfonic acid, p-CH3C6H4SO3H (Hpts; white crystalline
solid), lithium carbonate, Li2CO3, and lithium chloride were all
from Aldrich, as were methylsulfonic acid (CH3SO3H) and
sodium methylsulfonate (NaCH3SO3). Solutions of Lipts were
prepared by Li2CO3 neutralisation of Hpts and recrystallisation
from water (solubility ∼4 M). Lanthanum p-toluenesulfonate,
La(pts)3, was prepared by neutralisation of La2O3 (50 g; Rare-
earth Products Ltd., 99.5%) with aqueous 4.0 M Hpts at ∼80
ЊC, until the equivalent point is reached (pH ∼4).11 The solution
was then filtered, the white product recrystallised three times
from water and washed with acetone. A solid sample of the
CoIII oxidant NH4[Co(dipic)2]ؒH2O (dipic = 2,6-dicarboxyl-
atopyridine), which shows a UV-Vis peak at 510 nm (ε =
heterometallic derivative of [Mo3S4(H2O)9]4ϩ 2–4
.
Reaction with
Hg metal occurs readily, with the formation of the intensely
coloured purple double cube [Mo6HgS8(H2O)18]8ϩ 2
In 4 M
HCl, the double cube changes colour (purple blue), giving a
.
tetrachloro-substituted derivative, [Mo6HgS8(H2O)14Cl4]4ϩ, the
crystal structure of which has also been determined as a com-
ponent of a cucurbituril supramolecular assembly.3,4 Details of
the preparation of the Cd-containing double cubes [Mo6CdS8-
(dtp)8(CH3CN)2] (dtp = diethyl-dithiophosphate) and [Mo6Cd-
S8(Hnta)6]4Ϫ (nta = nitrilo-triacetate) from non-aqueous and
neutral aqueous solutions, respectively, have been reported,5
but no Zn-containing derivative has yet been identified. A
Cd-containing single cube is of considerable interest, since the
group 12 elements border the B and C categories previously
defined.1 Hexaaquacadmium() has an extensive solution chem-
istry.6 Aqueous solutions of 10Ϫ3 mM CdI2 have recently been
prepared.7
630 MϪ1 cmϪ1), was prepared as described previously.12
A
reduction potential for the [Co(dipic)2]Ϫ/2Ϫ couple (0.362 V vs.
NHE) has been determined.13
Instrumentation and techniques
A Phasesep Model LC2 chromatograph complete with thermal
conductivity detector (current 120 mA) and a 5 Å molecular
sieve was used for the quantitative determination of H2. Metal
analyses were carried out by inductively coupled plasma atomic
emission spectroscopy (ICP-AES) using an ATI Unicam Model
701 instrument. Applied Photophysics SX-17MV stopped-flow
and Perkin-Elmer Lambda 9 UV-Vis spectrophotometers were
used. For kinetic studies, ionic strengths were adjusted to I =
2.0 M with Lipts.
Experimental
Preparation of [Mo3S4(H2O)9]4؉
Stock solutions of the green MoIV incomplete cube [Mo3S4-
3
(H2O)9]4ϩ in 2 M HCl or 2 M Hpts were prepared from poly-
meric {Mo3S7Br4}x, as previously described.8,9 Details of the
UV-Vis absorption peak positions are λ/nm (ε/MϪ1 cmϪ1 per
Mo3) 370 (5.0 × 103), 616 (330) in 2 M HCl and 366 (5.5 × 103),
DOI: 10.1039/b109968j
J. Chem. Soc., Dalton Trans., 2002, 1941–1945
This journal is © The Royal Society of Chemistry 2002
1941

View Article Online
Treatment of kinetic data
ICP-AES. For this technique, solutions in HCl are generally
preferred to Hpts. However, since the cube readily decays in
HCl, a methylsulfonate solution was used here. The solution
eluted with 1.9 M CH3SO3Na ϩ 0.1 M CH3SO3H gave a ratio
Mo : Cd of 3.0 to 0.8. This suggests some decay of the Cd
product during elution, with loss of Cd2ϩ. The UV-Vis spec-
trum of a solution of Mo3CdS44ϩ in 1.9 M Lipts ϩ 0.1 M Hpts
was recorded. After reaction with O2, the final spectrum was
identified as [Mo3S4(H2O)9]4ϩ, with peaks at 366 and 603nm
(details as above). From this information, ε values for Mo3-
CdS44ϩ were calculated, Fig. 1, which has peak positions at λ/nm
Errors in kinetic data were obtained by unweighted least-
squares fitting procedures.
Results
Preparation of [Mo3CdS4(H2O)12]4؉
In the first procedure, stock [Mo3S4(H2O)9]4ϩ (2.5 mM; 40 mL)
in 2.0 M HCl was diluted 4-fold and CdCl2 (0.2 g, 0.001 mmol;
BDH) added prior to making the solution O2-free by bubbling
N2 through for ∼30 min. An excess of hypophosphorous acid
reductant (0.3 mL of a 50% solution; Aldrich) was added, and
the solution left at 20 ЊC for ∼1 h. Larger amounts of H3PO2
bring about precipitation. The colour of the solution changed
from green to orange–brown (eqn. 1).
4ϩ
Mo3S44ϩ ϩ Cd2ϩ ϩ 2eϪ
Mo3CdS4
(1)
Any reaction of the product with Hϩ at this stage results in a
regeneration of [Mo3CdS4(H2O)9]4ϩ, since Cd2ϩ and H3PO2 are
present in excess. The product solution was diluted to 0.3 M
HCl and loaded onto an ice-cooled Dowex 50W-X2 column
under N2. After washing with 0.10 M Hpts (100 mL) to remove
excess H3PO2, washing was continued with 0.10 M Lipts ϩ
0.10 M Hpts (150 mL) and 0.40 M Lipts ϩ 0.10 M Hpts
(150 mL) to remove excess Cd2ϩ. Elution of the orange–brown
[Mo3CdS4(H2O)12]4ϩ was initially by the displacement method
using 0.30 M La(pts)3 ϩ 0.10 M Hpts.11 However, because
quantitative studies were more difficult in the presence of the
3 :1 electrolyte, this procedure was superceded by one using a
solution of 2.9 M Lipts and 0.10 M Hpts for elution. Under
these conditions, no [Mo3S4(H2O)9]4ϩ band was observed and
yields approaching 100% were obtained. In the presence of
higher concentrations of Hϩ and with the addition of ClϪ,
a decay process yielding [Mo3S4(H2O)9]4ϩ and H2 becomes
increasingly important.
Fig. 1 UV-Vis spectrum of [Mo3CdS4(H2O)12]4ϩ in 1.90 M Lipts ϩ
0.10 M Hpts aqueous solution. The inset shows an enlargement of the
500–850 nm region.
(ε = MϪ1 cmϪ1 per Mo3) 374 (4580); 554sh (331); 866 (420). A
variety of procedures were used in attempts to obtain crystals.
The sensitivity of Mo3CdS44ϩ to Hϩ precludes the use of cucur-
bituril (soluble in 1–2 M HCl) to induce crystallisation.3
Assuming that the Cd is six-coordinate, the formula [Mo3Cd-
S4(H2O)12]4ϩ applies.
In a second procedure, Cd granules (1 g; 30–80 mesh;
Aldrich) were added to [Mo3S4(H2O)9]4ϩ in 0.50 M Hpts and
the solution heated (∼70 ЊC) for 1 h to give an orange–brown
solution (eqn. 2).
Reaction with [Co(dipic)2]؊
The stoichiometry was determined by titrating 0.10 mL
aliquots (Hamilton micro-syringe) of [Co(dipic)2]Ϫ (2.21 ×
10Ϫ3 M) into a solution of the Mo3CdS44ϩ cube (1.60 mL; 5.0 ×
4ϩ
4ϩ
Cd ϩ Mo3S4
Mo3CdS4
(2)
The Cd cube was again purified by Dowex chromatography
and eluted with 2.9 M Lipts ϩ 0.1 M Hpts. UV-Vis spectra
indicated 100% conversion to Mo3CdS44ϩ. With this procedure,
possible interference from H3PO2 and chloride (which can
affect kinetic studies) is avoided.
10Ϫ4 M), both in 1.90 M Lipts ϩ 0.10 M Hpts. A plot of
4ϩ
the absorbance at 866 nm against [Co(dipic)2Ϫ]/[Mo3CdS4
]
(inset to Fig. 2) gave a ratio of 2.08 at the end point. This
indicates a 2 : 1 stoichiometry (eqn. 3) with ∼4% uncertainty in
[Mo3CdS44ϩ].
4ϩ
Solutions of Mo3CdS4 stored at ∼4 ЊC under air-free con-
ditions (N2) age quickly, with deposition of a fine brown powder.
It is necessary to prepare fresh solutions after 2 days.
Attempts to prepare a corner-shared double cube
Conditions were as used previously for the conversion of the
single cube, e.g. [Mo3SnS4(H2O)12]6ϩ to [Mo6SnS8(H2O)18]8ϩ 10
. A
solution of [Mo3CdS4(H2O)12]4ϩ (3 mM; 5 mL) in 1.9 M
Lipts ϩ 0.1 M Hpts was mixed with [Mo3S4(H2O)9]4ϩ (3 mM; 5
mL) in 0.1 M Hpts and excess H3PO2 (1 mL of a 50% w/w H2O
solution), added under O2-free conditions. The mixture was left
for ∼1 h. No colour change was observed and on carrying out
Dowex chromatography, only the starting materials were
identified.
Stability in air and characterisation
On exposure of a solution of the cube in 1.9 M Lpts ϩ 0.1 M
Hpts to air with some shaking, ∼60% decay is observed in
Fig. 2 Rate constants for the [Co(dipic)2]Ϫ oxidation of [Mo3CdS4-
(H2O)12]4ϩ in aqueous 1.9 M Lipts ϩ 0.10 M Hpts solution. The inset
illustrates absorbance changes on titrating [Co(dipic)2]Ϫ from a micro-
syringe into [Mo3CdS4(H2O)12]4ϩ, I = 2.00 M (Lipts).
5 min, with the reformation of [Mo3S4(H2O)9]4ϩ and Cd2ϩ
.
Storage was under N2 at 4 ЊC and all experiments were carried
out in O2-free conditions. Metal analyses were carried out by
1942
J. Chem. Soc., Dalton Trans., 2002, 1941–1945

View Article Online
which gives kH = 4.6(2) × 10Ϫ4 MϪ2 Ϫ1. Smaller contributions
s
Mo3CdS44ϩ ϩ 2CoIII
Mo3S44ϩ ϩ Cd2ϩ ϩ 2CoII (3)
from a first-order [Hϩ] term cannot be ruled out, where a fit
to the dependence kobs = a[Hϩ]2 ϩ b[Hϩ] was tested by plotting
kobs/[Hϩ] vs. [Hϩ]. This gives a ≈ 2.8 × 10Ϫ4 MϪ2 sϪ1 and b ≤
0.80 × 10Ϫ4 MϪ1 sϪ1 (uncertainty in extrapolation), i.e. the first-
order rate constants for the reactions carried out are dominated
(∼80%) by an [Hϩ]2-dependent term.
One equivalent of [Mo3S4(H2O)9]4ϩ is released, confirming the
charge on [Mo3CdS4(H2O)12]4ϩ as 4ϩ.
The kinetics were monitored by studying the absorbance
decay of the cube at 866 nm by stopped-flow spectrophoto-
metry. First-order rate constants kobs were obtained for the
4ϩ
oxidation of Mo3CdS4 (∼0.15 mM) with [Co(dipic)2]Ϫ (1.8–
Reaction of [Mo3CdS4(H2O)12]4؉ with H؉ in the presence of Cl؊
6.7 mM). The [Co(dipic)2]Ϫ concentration was varied at [Hϩ] =
0.10 M, I = 2.00 M (Lipts). A ≥ 20-fold excess of oxidant
The reaction with Hϩ is faster in the presence of added LiCl.
The dependence on [ClϪ] was explored with [Mo3CdS4-
(H2O)124ϩ] in the range 0.23–0.48 mM, [ClϪ] = 0.5–1.5 M, I =
2.00 M (Lipts). At [Hϩ] ≤ 0.20 M, contributions from eqn. 5 are
small. The variation of kobs with [ClϪ] (Fig. 4) indicates a dom-
(≥ 3.6mM) is required to meet pseudo-first-order conditions for
eqn. 3. From Fig. 2, a first-order dependence on [Co(dipic)2
applies, with the rate law as in eqn. 4.
Ϫ
]
Rate = kCo[Mo3CdS44ϩ][CoIII]
(4)
The slope gives a second-order rate constant kCo = 31.5(5)
M
Ϫ1 sϪ1. From two runs at a higher [Hϩ] of 0.20 M, no depend-
ence of kCo on [Hϩ] was observed. This suggests an outer-
sphere reaction as in the reactions of other cube derivatives of
[Mo3S4(H2O)9]4ϩ, all of which give [Hϩ]-independent rate laws
over a wider range of [Hϩ] (0.5–2.0 M).1,8,10
Reaction of [Mo3CdS4(H2O)12]4؉ with H؉
Hydrogen gas is released and was identified and determined
quantitatively by gas chromatography. A solution of [Mo3Cd-
S4(H2O)12]4ϩ (2.09 mM; 15 mL) in 1.00 M Lipts ϩ 0.10 M Hpts
in a three-necked flask was made O2-free by passing N2 gas
through at 15.4 mL minϪ1 for ∼20 min. The [Hϩ] was adjusted
to 1.00 M and [ClϪ] = 0.9 M, I = 2.00 M. The solution was
stirred and heated to ∼50 ЊC. A brown to green colour change
was observed. The evolution of H2 was monitored by sampling
the N2 stream after 1 min, and every subsequent 2 min for
30 min. Calibration was using a 1.03% H2 sample. The amount
of H2 evolved was 3.110 × 10Ϫ5 mol (0.6969 mL), which
compares with the calculated amount of 3.135 × 10Ϫ5 mol from
eqn. 5.
Fig. 4 The variation of first-order rate constants kobs (25 ЊC) for the
reaction of Hϩ with [Mo3CdS4(H2O)12]4ϩ in the presence of [ClϪ], [Hϩ] =
0.20 M.
inant [ClϪ]2 term. This at first suggests ClϪ complexing to the
Cd, but an alternative interpretation is possible, as considered
below. From exploratory experiments, the [ClϪ]2 term has a
dependence on [Hϩ] in the range 0.10–0.20 M, with an [Hϩ]2
term again dominant.
Mo3CdS44ϩ ϩ 2Hϩ
Mo3S44ϩ ϩ H2 ϩ CdII
(5)
Discussion
In these studies, a cadmium single cube derivative of [Mo3-
Kinetic studies were carried out on the reaction of [Mo3Cd-
S4(H2O)12]4ϩ (0.188–0.383 mM), in ptsϪ solutions, [Hϩ] = 0.5–
1.5 M, I = 2.00 M (Lipts), by conventional UV-Vis monitoring
at 866 nm. Care was required to obtain reproducible data due
to the aging of stock solutions. First-order rate constants kobs
(25 ЊC) were determined from the slope of plots of absorbance
(A) changes, ln(At Ϫ A∞) vs. time. The inset to Fig. 3 suggests a
dependence of kobs vs. [Hϩ]2, as in eqn. 6
S4(H2O)9]4ϩ has been prepared for the first time. From ICP-AES
8ϩ
analyses, the double cube structure Mo6CdS8 (6 : 1 ratio
Mo : Cd) is ruled out, and the 3 : 1 ratio reported suggests a
core structure Mo3CdS44ϩ. The 2 : 1 stoichiometry (eqn. 3) for
the oxidation with [Co(dipic)2]Ϫ confirms the 4ϩ charge. The
cube is also oxidised by O2 (a common occurrence), and by Hϩ
(which is relatively rare); in all three redox processes,
[Mo3S4(H2O)9]4ϩ and Cd2ϩ are formed. In the reaction with Hϩ
(eqn. 5), replacement of ptsϪ with ClϪ produces a significant
increase in the rate. Following the recently adopted formalism,
the cube is written as Mo3S44ϩ,Cdo.14 The coordination number
of the Cd has not been determined, but is assumed to be octa-
hedral, which allows coordination of Hϩ, and possibly also ClϪ,
to take place.
kobs = kH [Hϩ]2
(6)
The rate law (eqn. 4) for the 2 : 1 [Co(dipic)2]Ϫ oxidation of
[Mo3CdS4(H2O)12]4ϩ indicates that the reaction with the first
[Co(dipic)2]Ϫ is rate determining. In Table 1, rate constants are
compared with those of other cubes,15–17 and Mo3CdS44ϩ is seen
to be mid-range in terms of redox reactivity. It has been shown
that the group 13 cubes [Mo3GaS4(H2O)12]5ϩ and [Mo3In-
S4(H2O)12]5ϩ are ∼20 times more reactive with [Co(dipic)2]Ϫ in
2.0 M HCl as compared to 2.0 M Hpts,17 while group 13 and 14
double cubes are ∼150 times more reactive.18 The most reason-
able explanation is that in 2.0 M ptsϪ, outer-sphere association
occurs with hydrogen bonding of adjacent ptsϪ ions, giving a
protective sheath around the cluster, as observed in the crystal
structures.18 In the case of [Mo3CdS4(H2O)12]4ϩ, addition of ClϪ
increases the rate of reaction with Hϩ. Coordination of ClϪ to
Fig. 3 First-order rate constants kobs (25 ЊC) for the reaction of
Hϩ with [Mo3CdS4(H2O)12]4ϩ (∼0.3 mM) and the [Hϩ]2 dependence, I =
2.00 M (Lipts).
J. Chem. Soc., Dalton Trans., 2002, 1941–1945
1943

View Article Online
Table 1 A comparison of rate constants (25 ЊC) for the outer-sphere
oxidation of some heterometallic derivatives of [Mo3S4(H2O)9]4ϩ in 2.0
M (Lipts), except where otherwise stated
Electrochemical and kinetic evidence has been obtained for
protonation of µ-S groups of Fe–S clusters.26,27 As yet, how-
ever, no UV-Vis spectrophotometric evidence for protonation
of µ-S groups in Mo–S clusters has been obtained.1 In studies
on [3Fe–4S] clusters in Desulfovibrio africanus, uptake of
heterometals (M = Fe, Zn, Co and Cd) has been observed.26
Cluster
kCo/MϪ1 sϪ1
Ref.
[Mo3CdS4(H2O)12]4ϩ
[Mo3GaS4(H2O)12]5ϩ
[Mo3InS4(H2O)12]5ϩ
[Mo3FeS4(H2O)10]4ϩ
[Mo3NiS4(H2O)10]4ϩ
[Mo6InS8(H2O)18]8ϩ
[Mo6InO2S6(H2O)18]8ϩ
[Mo6TlS8(H2O)18]8ϩ
31.5
0.23
0.65
87a
This work
o
2ϩ
Reversible interconversion of Fe3S4 and Fe3CdS4 occurs on
17
17
15
16
17
17
17
addition of Cd2ϩ, while Fe3S4 and the hyper-reduced (all
ϩ
FeII) cluster Fe3S4 have little affinity for M. In studies27 on
2Ϫ
0.35a
30.9
322
the oxidation of Fe3S42Ϫ, a net uptake of three protons (pHs
at or close to 7.0) has been reported. No H2 evolution is
observed in any of these, in contrast to hydrogenase enzymic
reactions.28
fast
a I adjusted to 2.0 M with LiClO4.
A further question is concerned with the role of ClϪ and
whether a step involving a direct interaction with the Cd is
relevant instead of (or alongside) the ClϪ vs. ptsϪ effect already
considered. The most obvious argument in favour of ClϪ cat-
alysis is based on electrostatics, with Hϩ more readily able to
approach the Cd when it is coordinated to chloride. Equi-
librium constants (K/MϪ1) for the formation of monochloro
complexes at the heterometal of [Mo3MS4(H2O)10]4ϩ have been
reported for M = Fe (560),15 and M = Ni (97).16 The main group
metal Sn in [Mo3SnS4(H2O)12]6ϩ has an even greater affinity for
ClϪ, with the formation of [Mo3(SnCl3)S4(H2O)9]3ϩ approach-
[Cd(H2O)6]2ϩ is more favourable by an order of magnitude as
compared to other transition metal ions.19
The oxidation of heterometallic derivatives of [Mo3S4-
(H2O)9]4ϩ with Hϩ has been observed previously,17 but is a rela-
tively rare occurrence confined to group 12 and 13 heteroatoms.
Evolution of H2 gas has been monitored in all cases by gas
chromatography.17 Reactions of the corner-shared double cubes
[Mo6TlS8(H2O)18]8ϩ (kH
= 0.25 × ) and
10Ϫ3 MϪ1 sϪ1
[Mo6InO2S6(H2O)18]8ϩ (kH = 4.9 × 10Ϫ3 MϪ1 sϪ1) with [Hϩ] have
been studied in the range 0.5–2.0 M, I = 2.00 M (LiCl). In both
cases, a first-order dependence on [Hϩ] is observed. A similar
ing completion at 0.05 M ClϪ.11 Complexation of ClϪ to the
29
Mo atoms of [Mo4S4(H2O)12]5ϩ (K
=
1.98 MϪ1
)
and
[Mo3S4(H2O)9]4ϩ (K = 3.0 MϪ1) is much less at 25 ЊC, I = 2.00 M
(LiClO4).30 Overall however, in view of the earlier comments, it
would be surprising if ClϪ complexation to the heteroatom
made other than minor contributions.
decay process is observed with [Mo6InS8(H2O)18]8ϩ 17
. The reac-
tion of [Mo6TlS8(H2O)18]8ϩ with Hϩ occurs in a single stage
(eqn. 7)
To summarise, the single cube [Mo3CdS4(H2O)12]4ϩ has been
prepared and characterised. Oxidation decay processes are
observed with both O2 and Hϩ, when [Mo3S4(H2O)9]4ϩ and
Cd2ϩ are formed. Rate constants for the oxidation of [Mo3Cd-
S4(H2O)12]4ϩ with [Co(dipic)2]Ϫ are mid-range (31.5 MϪ1 sϪ1) for
heterometallic derivatives of [Mo3S4(H2O)9]4ϩ. The reaction
with Hϩ has novel features, in particular the dominant [Hϩ]2
dependence of the rate law. In the presence of ClϪ for ptsϪ the
cube is more exposed to direct attack by Hϩ; half-life in 1 M
HCl ∼5 min. The inhibition by ptsϪ is attributed to its tendency
to form a protective outer sphere layer. The reactions of the Cd,
Ga, In and Tl heterometallic derivatives of [Mo3S4(H2O)9]4ϩ
with Hϩ provides an interesting development in the chemistry
of this type of cube.
Mo6TlS88ϩ ϩ Hϩ
2Mo3S44ϩ ϩ Tlϩ ϩ 1/2H2
(7)
and the reaction of [Mo6InS8(H2O)18]8ϩ occurs in two steps
(eqn. 8 and eqn. 9).17
Mo6InS88ϩ ϩ Hϩ
Mo3InS45ϩ ϩ Mo3S44ϩ ϩ 1/2H2 (8)
Mo3S44ϩ ϩ In3ϩ ϩ H2
(9)
Mo3InS45ϩ ϩ 2Hϩ
The reaction of the Inϩ aqua ion with Hϩ gives a first-order
dependence on [Hϩ].20,21 In all these reactions, the behaviour
observed contrasts with the dominant [Hϩ]2 rate law term
observed for eqn. 10.
Acknowledgements
Mo3CdS44ϩ ϩ 2Hϩ
Mo3S44ϩ ϩ Cd2ϩ ϩ H2 (10)
We are grateful for financial support from the University of
Newcastle, and leave of absence from the Institute of Inorg-
anic Chemistry, RAS, Novosibirsk, Russia (M. N. S.) and the
Department of Inorganic Chemistry of La Laguna, Tenerife,
Spain (R. H.-M.).
It is probably no coincidence that the hydride chemistry of
group 12 and 13 metals has been investigated.22 The hydride
ZnH2 23 is more stable than CdH2 and HgH2, which decompose
rapidly, even below 0 ЊC.24 The formation of InHCl2, InH2Cl
and the gallium analogues has also been reported.25 In the pres-
ent studies, the formation of a dihydride by the reaction of two
Hϩ with the Cd of Mo3CdS44ϩ is possible. The effect of ClϪ can
be summarised as a dominant [ClϪ]2 dependence, which may, in
some part, be due to inner-sphere complexation to the Cd.
However, in view of the ptsϪ vs. ClϪ effects already noted for the
group 13 and 14 cubes with [Co(dipic)2]Ϫ,17,18 inhibition by
hydrogen-bonded outer-sphere ptsϪ ions is a more likely inter-
pretation. Such a protective sheath is a common feature in crys-
tal structures of ptsϪ salts.3 In solution, on replacing ClϪ with
ptsϪ, the inhibition of the reaction of the heterometallic cube
with Hϩ (or [Co(dipic)2]Ϫ) is accounted for by this decreased
access. There are two different ways in which an activated com-
plex of [Mo3CdS4(H2O)12]4ϩ and two Hϩ ions can be assembled.
Both Hϩ can coordinate to the Cd as hydrido groups. Altern-
atively, direct attack of Hϩ on an existing hydride ligand is
possible.
References
1 R. Hernandez-Molina, M. N. Sokolov and A. G. Sykes, Acc. Chem.
Res., 2001, 34, 223–230.
2 T. Shibahara, H. Akashi, M. Yamasaki and K. Hashimoto, Chem.
Lett., 1991, 689–692.
3 M. N. Sokolov, A. V. Virovets, D. N. Dybtsev, O. A. Gerásko, V. P.
Fedin, R. Hernandez-Molina, W. Clegg and A. G. Sykes, Angew.
Chem., Int. Ed., 2000, 39, 1659–1661.
4 V. P. Fedin, M. N. Sokolov, G. J. Lamprecht, R. Hernandez-Molina,
M.-S. Seo, A. V. Virovets, W. Clegg and A. G. Sykes, Inorg. Chem.,
2001, 40, 6598–6603.
5 G. Sakane, H. Kawasaki, M. Yamasaki, H. Adachi and T.
Shibahara, Chem. Lett., 1999, 631–635.
6 N. N. Greenwood and A. Earnshaw, Chemistry of the Elements,
Butterworth and Heineman, Oxford, 2nd edn., 1997, p. 1208.
7 O. A. Babich and E. S. Gould, Chem. Commun., 2001, 998.
8 M. N. Sokolov, N. Coichev, H. D. Moya, R. Hernandez-Molina,
C. D. Borman and A. G. Sykes, J. Chem. Soc., Dalton Trans., 1997,
1863–1870.
1944
J. Chem. Soc., Dalton Trans., 2002, 1941–1945

View Article Online
9 D. M. Saysell, V. P. Fedin, G. J. Lamprecht, M. N. Sokolov and
A. G. Sykes, Inorg. Chem., 1997, 36, 2982–2987.
10 R. Hernandez-Molina, D. N. Dybtsev, V. P. Fedin, M. R. J.
Elsegood, W. Clegg and A. G. Sykes, Inorg. Chem., 1998, 37, 2995–
3001.
11 J. E. Finholt, M. E. Thompson and R. E. Connick, Inorg. Chem.,
1981, 20, 4150–4155.
12 J. E. Varey, G. J. Lamprecht, V. P. Fedin, A. Holder, W. Clegg, M. R.
J. Elsegood and A. G. Sykes, Inorg. Chem., 1996, 35, 5525–5530.
13 C. G. Saysell, C. D. Borman, A. J. Baron, M. J. McPherson and
A. G. Sykes, Inorg. Chem., 1997, 36, 4520–4525.
19 (a) P. Gerding, Acta Chem. Scand., 1966, 20, 79; (b) P. Gerding and
I. Jönsson, Acta Chem. Scand., 1968, 22, 2247–2554.
20 R. S. Taylor and A. G. Sykes, J. Chem. Soc. A, 1969, 2419–2423.
21 K. Chandra and E. S. Gould, Inorg. Chem., 1996, 35, 3881–3884.
22 T. M. Greene, L. Andrews and A. J. Downs, J. Am. Chem. Soc.,
1995, 117, 8180–8187.
23 J. Watkins and E. C. Ashby, Inorg. Chem., 1974, 13, 2350–2354.
24 G. D. Barbaras, C. Dillard, A. E. Finholt, J. Wartick and H. S.
Schlesinger, J. Am. Chem. Soc., 1951, 73, 4584–4588.
25 H.-J. Himmel, A. J. Downs and T. M. Greene, J. Am. Chem. Soc.,
2000, 122, 922–930.
14 R. Hernandez-Molina, A. J. Edwards, W. Clegg and A. G. Sykes,
Inorg. Chem., 1998, 37, 2989–2994.
15 P. W. Dimmock, D. P. E. Dickson and A. G. Sykes, Inorg. Chem.,
1990, 29, 5120–5125.
16 D. M. Saysell, C. D. Borman, C.-H. Kwak and A. G. Sykes, Inorg.
Chem., 1996, 35, 173–176.
26 J. N. Butt, S. E. J. Fawcett, J. Breton, A. J. Thomson and
F. A. Armstrong, J. Am. Chem. Soc., 1997, 119, 9729–9737.
27 J. L. C. Duff, J. L. J. Breton, J. N. Butt, F. A. Armstrong and A. J.
Thomson, J. Am. Chem. Soc., 1996, 118, 8593–8603.
28 e. g. J. C. Fontecilla-Camps and S. W. Ragsdale, Adv. Inorg. Chem.,
1999, 47, p. 283–333.
17 R. Hernandez-Molina, V. P. Fedin, M. N. Sokolov, D. M. Saysell
and A. G. Sykes, Inorg. Chem., 1998, 37, 4328–4334.
18 M.-S. Seo, V. P. Fedin, R. Hernandez-Molina, M. N. Sokolov,
A. Sokolowski, M. R. J. Elsegood, W. Clegg and A. G. Sykes, Inorg.
Chem., 2001, 40, 6115–6122.
29 Y.-J. Li, M. Nasreldin, M. Humanes and A. G. Sykes, Inorg. Chem.,
1992, 31, 3011–3017.
30 D. T. Richens, P.-A. Pittet, A. E. Merbach, M. Humanes, G. J.
Lamprecht, B.-L. Ooi and A. G. Sykes, J. Chem. Soc., Dalton Trans.,
1993, 2305–2311.
J. Chem. Soc., Dalton Trans., 2002, 1941–1945
1945