Reversible capture of small molecules on bimetallaborane clusters: Synthesis, structural characterization, and photophysical aspects

Article abstract of DOI:10.1021/ic200374k

Metallaborane compounds containing two adjacent metal atoms, [(PMe ₂Ph)₄MM′B₁₀H₁₀] (where MM′ = Pt₂, 1; PtPd, 7; Pd₂, 8), have been synthesized, and their propensity to sequester O₂, CO, and SO ₂ and to then release them under pulsed and continuous irradiation are described. Only [(PMe₂Ph)₄Pt₂B ₁₀H₁₀], 1, undergoes reversible binding of O₂ to form [(PMe₂Ph)₄(O₂)Pt₂B ₁₀H₁₀] 3, but solutions of 1, 7, and 8 all quantitatively take up CO across their metal-metal vectors to form [(PMe₂Ph) ₄(CO)Pt₂B₁₀H₁₀] 4, [(PMe ₂Ph)₄(CO)PtPdB₁₀H₁₀] 10, and [(PMe₂Ph)₄(CO)Pd₂B₁₀H₁₀] 11, respectively. Crystallographically determined interatomic M-M distances and infrared CO stretching frequencies show that the CO molecule is bound progressively more weakly in the sequence {PtPt} > {PtPd} > {PdPd}. Similarly, SO₂ forms [(PMe₂Ph)₄(SO ₂)Pt₂B₁₀H₁₀] 5, [(PMe ₂Ph)₄(SO₂)PtPdB₁₀H₁₀] 12, and [(PMe₂Ph)₄(SO₂)Pd₂B ₁₀H₁₀] 13 with progressively weaker binding of the SO ₂ molecule. The uptake and release of gas molecules are accompanied by changes in their absorption spectra. Nanosecond transient absorption spectroscopy clearly shows that the O₂ and CO molecules are liberated from the bimetallic binding site with high quantum yields of about 0.6. For 3, in addition to dioxygen release in the triplet ground state, singlet oxygen O₂(¹Δ_g) was also detected with a quantum yield <0.01. In most cases, the release and rebinding of the gas molecules can be cycled with little photodegradation of the compounds. Femtosecond transient absorption spectroscopy further reveals that the photorelease of the O₂ and CO molecules, from 3 and 4 respectively, is an ultrafast process taking place on a time scale of tens of picoseconds. For SO₂, the release is even faster (<1 ps), but only in the case of mixed metal PtPd adducts, most probably because of the metal-metal bonding asymmetry in the mixed metal clusters; for the corresponding symmetric Pt₂ and Pd ₂ adducts, 5 and 13, the release of SO₂ is significantly slower (>1 ns). All these compounds may have potential to serve as light-triggered local and instantaneous sources of the studied gases.

Full text of DOI:10.1021/ic200374k

ARTICLE
pubs.acs.org/IC
Reversible Capture of Small Molecules On Bimetallaborane Clusters:
Synthesis, Structural Characterization, and Photophysical Aspects
Jonathan Bould,*,†,^ Tomꢀaꢁs Baꢁse,† Michael G. S. Londesborough,† Luis A. Oro,‡ Ramꢀon Macías,‡
John D. Kennedy,^ Pavel Kubꢀat,§ Marcel Fuciman,|| Tomꢀaꢁs Polívka,|| and Kamil Lang*,†
†
ꢁ
Institute of Inorganic Chemistry, Academy of Sciences of the Czech Republic, v.v.i., 250 68 Husinec-Reꢁz, Czech Republic
‡Instituto Universitario de Catꢀalisis Homogꢀenea, Universidad de Zaragoza, 50009-Zaragoza, Spain
§J. Heyrovskꢀy Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, v.v.i., Dolejꢁskova 3, 182 23 Prague 8,
Czech Republic
Institute of Physical Biology, University of South Bohemia, Zꢀamek 136, 373 33 Novꢀe Hrady, Czech Republic
^The School of Chemistry of the University of Leeds, Leeds LS2 9JT, U.K.
S Supporting Information
b
ABSTRACT: Metallaborane compounds containing two adjacent metal
atoms, [(PMe2Ph)4MM0B10H10] (where MM0 = Pt2, 1; PtPd, 7; Pd2, 8), have
been synthesized, and their propensity to sequester O2, CO, and SO2 and
to then release them under pulsed and continuous irradiation are described.
Only [(PMe2Ph)4Pt2B10H10], 1, undergoes reversible binding of O2 to form
[(PMe2Ph)4(O2)Pt2B10H10] 3, but solutions of 1, 7, and 8 all quantitatively
take up CO across their metalꢀmetal vectors to form [(PMe2Ph)4-
(CO)Pt2B10H10] 4, [(PMe2Ph)4(CO)PtPdB10H10] 10, and [(PMe2Ph)4(CO)-
Pd2B10H10] 11, respectively. Crystallographically determined interatomic MꢀM
distances and infrared CO stretching frequencies show that the CO molecule is
bound progressively more weakly in the sequence {PtPt} > {PtPd} > {PdPd}.
Similarly, SO2 forms [(PMe2Ph)4(SO2)Pt2B10H10] 5, [(PMe2Ph)4(SO2)-
PtPdB10H10] 12, and [(PMe2Ph)4(SO2)Pd2B10H10] 13 with progressively
weaker binding of the SO2 molecule. The uptake and release of gas molecules are accompanied by changes in their absorption
spectra. Nanosecond transient absorption spectroscopy clearly shows that the O2 and CO molecules are liberated from the
bimetallic binding site with high quantum yields of about 0.6. For 3, in addition to dioxygen release in the triplet ground state, singlet
oxygen O2(1Δg) was also detected with a quantum yield <0.01. In most cases, the release and rebinding of the gas molecules can be
cycled with little photodegradation of the compounds. Femtosecond transient absorption spectroscopy further reveals that the
photorelease of the O2 and CO molecules, from 3 and 4 respectively, is an ultrafast process taking place on a time scale of tens of
picoseconds. For SO2, the release is even faster (<1 ps), but only in the case of mixed metal PtPd adducts, most probably because of
the metalꢀmetal bonding asymmetry in the mixed metal clusters; for the corresponding symmetric Pt2 and Pd2 adducts, 5 and 13,
the release of SO2 is significantly slower (>1 ns). All these compounds may have potential to serve as light-triggered local and
instantaneous sources of the studied gases.
’ INTRODUCTION
orange dioxygen bimetallaborane complex [(PMe2Ph)4(O2)-
Pt2B10H10] 3 (Scheme 1).1 The reversibility is readily induced
by pressure reduction or by mild warming. Also, by analogy with
the competitive binding of gases other than oxygen to hemoglo-
bin, compound 1 interacts with CO and SO2 to give tan-colored
[(PMe2Ph)4(CO)Pt2B10H10] 4 and orange [(PMe2Ph)4(SO2)-
Pt2B10H10] 5 (Scheme 1).1b In a similar manner to O2, the up-
take of CO and SO2 by 1 occurs very readily and quantitatively,
but in these cases the additions are not reversible under normal
conditions of pressure reduction or mild heating. Reversible SO2
The reversible binding of small molecules on metal atoms,
metalꢀmetal vectors, and metal surfaces is often used, both
in Nature and in industry, to overcome otherwise insurmoun-
table thermodynamic barriers in chemistry. Examples of uses
found for this type of interaction are numerous and perhaps
the most important one is the reversible uptake of dioxygen
on the iron center in hemoglobin. In this context, we have
previously reported the quantitative and reversible uptake of
O2 by the metalꢀmetal linkage in the intensely dark purple
icosahedral twelve-vertex bimetallaborane cluster compound
[(PMe2Ph)4Pt2B10H10] 1 made from the reaction of [PtCl2-
(PMe2Ph)2] with [(PMe2Ph)2PtB10H12] 2 to give the yellow-
Received: February 22, 2011
Published: July 21, 2011
r
2011 American Chemical Society
7511
dx.doi.org/10.1021/ic200374k Inorg. Chem. 2011, 50, 7511–7523
|

Inorganic Chemistry
ARTICLE
Scheme 1. Schematic Structures of Investigated Compounds
Showing Compound Numbering and Acronyms Used in the
Text
Figure 1. Representation of the crystallographically obtained molecular
structure of [(PMe2Ph)4Pd2B10H10], Pd2 8, showing the cluster num-
bering system employed, with 50% thermal ellipsoids for non-hydrogen
atoms. The phenyl rings (except for the ipso carbon atoms) and methyl
group hydrogen atoms are omitted to aid clarity. The molecule possesses
a C2 axis such that equivalent cluster positions are Pd(1)tPd(2),
B(3)tB(6), B(4)tB(11), B(5)tB(7), B(8)tB(10), and B(9)tB-
(12). Selected measured geometric parameters for Pd2 8 are listed in
Table 1.
by filtration in yields of up to 76% and 57%, respectively, with
the latter, new compound, affording a satisfactory elemental
analysis. The yield for 1 is an improvement on the 23%
previously reported from preparations using tetrahydrofuran
(THF) as solvent.1b Interestingly, it may also be noted that a
crystalline precipitate of the Pt2 1 compound may also be
obtained in about 50% yield from a toluene solution of
[(PMe2Ph)2PtB10H12] at reflux containing added PMe2Ph9
(see Experimental Section). Single-crystal X-ray diffraction data
were obtained for the dipalladium compound 8 (Figure 1),
together with the cluster numbering scheme. The molecular
structure of Pt2 1 is previously reported.1 The isolation of the
mixed palladiumꢀplatinum analogue PtPd 7 from the reaction
mixture using KHBEt3 is less straightforward. Crystallization of
the purple-brown solid afforded in the reaction is difficult and
results in some decomposition of the compound. Nevertheless,
sufficient quantities for a satisfactory elemental analysis and for
further reaction chemistry have been obtained (31% yield). Its
identity is also readily inferred from the results of multinuclear
NMR spectroscopy in comparison with those for the Pt2 1 and
Pd2 8, (Supporting Information) and from its well-character-
ized adducts with SO2 and CO, which were made in situ as
described below.
and/or CO uptake on monometallic2 or bimetallic species3ꢀ6
has been reported previously, but rarely in conjunction with
oxygen uptake7 as occurs in the system we describe here. The
reversibility of O2 addition to 1 to give 3 contrasts with the
irreversible formation of the only other known analogous O2
complex to a metalꢀmetal-bonded system, that is, the formation
of dark purple [Ir2I2(CO)2(O2)(Ph2PCH2PPh2)2] 6 from [Ir2I2-
(CO)2(Ph2PCH2PPh2)2], first reported 20 years ago.8
In this paper, we investigate whether the sequestration of SO2
and CO by our bimetallic system can be made to be reversible by
changing one or both of the metal atoms from platinum to
palladium. Additionally, we demonstrate that sequestered mol-
ecules of O2, CO, and SO2 can be reversibly ejected by pulsed
UVꢀvis irradiation: here, time-resolved spectroscopic studies
spanning the time-range from femtoseconds to microseconds
provide a quantitative refinement of the qualitative observations
obtained from steady-state UVꢀvis spectroscopy.
As an easy reference, and as an aid to the reader, Scheme 1
shows a schematic listing of the structures described in this paper
and their compound numbers. For example, [(PMe2Ph)4(O2)-
Pt2B10H10] 3, is referred to below as Pt2-O2 3.
The diplatinum compound 1 was first isolated in 1985,9
though its propensity to take up O2 was not recognized until
some 20 years later.1a The dark-purple crystalline solid exhibits
excellent air stability; a small sample of the compound, stored in a
sample tube with plastic cap, was found to be unchanged after 20
years. The two new species PtPd 7 and Pd2 8 are also air-stable in
the solid state, but samples of Pd2 8 have been observed to
degrade over a period of a few months. They are, however, stable
in chlorinated solvents in the absence of air. In the presence of air,
solutions of Pd2 8 in CHCl3 are not stable, perhaps suggesting
activation of atmospheric oxygen via a transient weak dioxygen
’ RESULTS AND DISCUSSION
Compound Syntheses. The bimetallic compounds [(PMe2Ph)4-
MM0B10H10] are synthesized by the addition of 2 equivalents of
KHBEt3 to suspensions of [(PMe2Ph)2MB10H12] (M = Pt or
Pd) in toluene at low temperature, followed by the addition of
[M0Cl2(PMe2Ph)2] (MM0 = PtPt for 1, PtPd for 7, and PdPd
for 8) and slow warming to room temperature. The air-stable
Pt2 1 and Pd2 8 compounds are then obtained essentially pure
7512
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
Table 1. Selected Measured Interatomic Dimensions from Single-Crystal X-ray Diffraction Studies with Selected Equivalent DFT
Calculated Dimensions in [Square Brackets]
Pt2
1a,c
Pd2
Pt2-CO
PtPd-CO
Pd2-CO
Pt2-SO2
PtPd-SO2
Pd2-SO2
8c
4a
10b
11
5a
12b
13
M(1)ꢀM(2)
2.9552(4)
[2.9545]
2.9163(3)
[2.9095]
2.7487(3)
[2.8246]
2.3402(16)
2.3516(17)
2.3331(17)
2.3440(17)
2.098(6)
[2.219]
2.749(3)
[2.7961]
2.3416(18)
2.3656(17)
2.378(3)
2.419(4)
2.112(4)
[2.260]
2.7491(6)
[2.785]
2.8194(4)
[2.8728]
2.383(3)
2.353(3)
2.345(2)
2.376(2)
2.388(2)
[2.5942]
2.355(2)
[2.5942]
2.339(10)
2.258(10)
2.219(12)
2.322(10)
2.301(9)
2.348(6)
2.258(10)
2.219(12)
1.454(7)
[1.490]
2.810(7)
[2.8545]
2.384(4)
2.351(4)
2.421(6)
2.379(7)
2.367(4)
[2.6465]
2.405(7)
[2.5592]
2.284(10)
2.232(9)
2.239(9)
2.328(10)
2.330(11)
2.329(11)
2.223(11)
2.229(12)
1.458(6)
[1.486]
2.8041(6)
[2.8375]
2.4172(13)
2.3798(13)
2.4276(15)
2.3915(13)
2.3689(13)
[2.5909]
2.4139(13)
[2.5909]
2.304(5)
2.260(5)
2.253(5)
2.313(5)
2.304(5)
2.324(5)
2.237(5)
2.262(6)
1.455(4)
[1.485]
M(1)ꢀP(1)
M(1)ꢀP(2)
M(2)ꢀP(3)
M(2)ꢀP(4)
M(1)ꢀC/S
2.3344(4)
2.3382(4)
2.3860(6)
2.3806(6)
2.4007(11)
2.4013(9)
2.4090(12)
2.4006(8)
2.148(3)
[2.249]
M(2)ꢀC/S
2.106(6)
[2.219]
2.167(5)
[2.203]
2.175(3)
[2.249]
M(1)ꢀB(3)
M(1)ꢀB(4)
M(1)ꢀB(5)
M(1)ꢀB(6)
M(2)ꢀB(3)
M(2)ꢀB(6)
M(2)ꢀB(7)
M(2)ꢀB(11)
S(1)\C(1)ꢀO(1)
2.3626(17)
2.2035(18)
2.2225(18)
2.346(3)
2.212(3)
2.233(3)
2.332(7)
2.264(7)
2.255(7)
2.305(6)
2.298(7)
2.348(7)
2.280(6)
2.266(6)
1.174(7)
[1.1676]
2.314(4)
2.268(4)
2.265(4)
2.333(4)
2.325(5)
2.291(5)
2.261(5)
2.268(5)
1.153(4)
[1.1623]
2.329(3)
2.293(3)
2.280(3)
2.305(3)
2.306(3)
2.344(3)
2.269(3)
2.284(3)
1.158(3)
[1.1581]
S(1)ꢀO(2)
O(1)ꢀS(1)ꢀO(2)
1.461(7)
113.7(4)
[115.7]
1.456(6)
112.7(4)
[115.8]
1.460(4)
113.5(2)
[116.1]
M(1)ꢀC(1)\S(1)ꢀM(2)
81.36(19)
[79.1]
79.94(15)
[77.6]
78.97(10)
[76.5]
72.94(6)
[67.2]
72.14(18)
[66.5]
71.78(4)
[66.4]
a Data from reference 1b. b M(1) = Pt. c There is crystallographically imposed 2-fold symmetry for compounds 1 and 8, such that M(1), P(1), P(2), B(3),
B(4), and B(5) are equivalent to M(2), P(3), P(4), B(6), B(11), and B(7), respectively.
adduct, although monitoring by 11B NMR spectroscopy shows
no sign of such an intermediate. We have not yet conducted a
detailed investigation of the products from these air-induced
reactivities, but we have established by a single-crystal X-ray
diffractions analysis that one product is a yellow-colored com-
pound [(Cl)(PMe2Ph)3Pd2B10H9(PMe2Ph)] 9 (Figure S1,
Supporting Information). Here, a PMe2Ph ligand has been
displaced from a palladium center by a chlorine substituent,
and the PMe2Ph has moved onto a boron atom. This results in a
more open crevice for access to the dipalladium center, which
augurs for some synthetic usefulness for this type of process, as
fine tailoring of these dimetal species for small-molecule seques-
tration and activation will involve both entropic adjustment of
the cavity size about the dimetal center and also electronic
adjustment by substituents on the boron atoms of the cluster.
The diplatinum analogue of 9 was isolated in the original work
that resulted in the Pt2 1 compound.9
Figure 2. Representation of the crystallographically determined mo-
lecular structure of [(PMe2Ph)4(CO)Pd2B10H10], Pd2-CO 11, dis-
Structural Studies of the CO and SO2 Complexes. We
played with 50% thermal ellipsoids for non-hydrogen atoms. Phenyl
initially assessed the system by DFT calculation of molecular
group atoms (with the exception of the ipso carbon atoms) and the
geometry at the B3LYP/6-31G*/LANL2DZ level10 to build up a
methyl group hydrogen atoms are omitted to aid clarity. Selected
measured geometric parameters are listed in Table 1.
matrix of information as to how modification of the metal would
be expected to weaken or strengthen the binding of the small
molecules, to validate our experimental observations thus en-
abling us to identify future synthetic targets. Details on these
preliminary predictive calculations may be seen in the Supporting
Information but, for comparison purposes, selected calculated
geometric dimensions for the molecules reported here are listed
7513
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
Figure 4. Measured (lower line) and DFT-calculated (upper line)
infrared stretching frequencies of the CO group in Pt2-CO 4 (right)
PtPd-CO 10 (center) and Pd2-CO 11 (left), plotted against the B3LYP/
6-31G*/LANL2DZ calculated CꢀO distance.
analogue are listed in Table 1. The overall structure of PtPd-
CO 10 is very similar to that for Pd2-CO 11, but the disordered
positions for the Pt and Pd atoms and for two of the PMe2Ph
groups means that the measured interatomic dimensions are less
precise than those for the Pt2-CO 4 and Pd2-CO 11, thus preclud-
ing a detailed comparison.
Sulfur dioxide reacts with PtPd 7 and Pd2 8 by addition to the
metalꢀmetal vector to give the two yellow-orange colored
species PtPd-SO2 12 and Pd2-SO2 13, respectively. The reaction
is also immediate and quantitative, as with Pt2 1, which gives light
orange Pt2-SO2 5.1b The compounds PtPd-SO2 12 and Pd2-SO2
13 are characterized by single-crystal X-ray diffraction analyses
(Figure 3, Table 1) and by NMR spectroscopy (Experimental
Section; a detailed analysis of the 11B NMR spectra of the new
compounds described in this paper may be also found in the
Supporting Information).
All three SO2 adducts, Pt2-SO2 5, PtPd-SO2 12, and Pd2-SO2
13, exhibit qualitatively very similar SO2 binding modes to the
metalꢀmetal unit. The sulfur atoms bridge the metalꢀmetal
linkage, which is not cleaved, and the sulfur atom exhibits
approximately tetrahedral coordination to the two metal
atoms and to the two oxygen atoms. In quantitative terms, with
allowance for the disorder in the crystallographic structure of
PtPd-SO2 12, the density functional theory (DFT) predictions
are nicely supported (Table 1), showing progressive lengthening
of the metal-to-sulfur distances and progressively decreasing
differences for the intermetal separations, from 2.8194(4)
[2.8728 calc.] for 5, to 2.8041(6) [2.8375 calc.] Å for 13. Some
minor differences are evident between the experimentally deter-
mined structure and the calculated structure, such as a tilting of
the SO2 group with respect to the plane defined by Pd(1)Pd-
(2)B(9)B(12), which is not evident in the calculated structure.
This may reasonably be ascribed to the effect of the packing of the
SO2 molecule within the PMe2Ph ligand sphere, which is itself
constrained by crystal packing forces. The organyl groups were
replaced by H atoms to reduce computation time and thus the
effect due to the constraints of the organyl groups would not
be seen.
Figure 3. Representation of the crystallographically determined mo-
lecular structure of [(PMe2Ph)4(SO2)Pd2B10H10], Pd2-SO2 13, dis-
played with 50% thermal ellipsoids for non-hydrogen atoms. Phenyl
group atoms, with the exception of the ipso carbon atoms and the methyl
group hydrogen atoms are omitted to aid clarity. The lower diagram
shows a side view illustrating the tilting of the SO2 group with respect to
the Pd(1)Pd(2)B(9)B(12) plane. The corresponding representation of
the molecular structure of the PtPd-SO2 12 analogue is visually
isostructural and is therefore not shown. Selected measured geometric
parameters for 12 and 13 are listed in Table 1.
in Table 1 and pertinent values shown in the Supporting Informa-
tion are referred to in the text as required.
In contrast to Pt2 1, no color changes are observed when
dichloromethane solutions of PtPd 7 or Pd2 8 are saturated with
dioxygen, and NMR spectroscopy shows no change. This reflects
the trend in the calculated OꢀO interatomic distances that
feature a shortening from 1.367 Å for Pt2-O2 to 1.351 Å for PtPd-
O2 to 1.336 Å for Pd2-O2 (Scheme S4, Supporting Information),
synonymous with a weakening the interaction between dioxygen
and the metalꢀmetal cluster vector.
Under one atmosphere pressure of CO at room temperature,
solutions of PtPd 7 and Pd2 8 exhibit immediate color changes
from brown and dark purple-red respectively, to a pale tan-yellow
in both cases. IR spectroscopy indicates the formation a CO
adduct [ν(CO) 1847 cmꢀ1 for PtPd-CO 10, 1882 cmꢀ1 for Pd2-
CO 11 (Pt2-CO 4 is 1810 cmꢀ1)1b]. This was confirmed by a
single-crystal X-ray diffraction studies on 10 and 11 (Figure 2)
and by 1H, 11B, and 31P NMR spectroscopy which show that the
addition is quantitative. Selected calculated and experimental
geometric parameters together with those of the Pt2-CO 4
The calculated intermetal distances are longer than the
measured ones by up to about 0.05 Å but this is not unexpected
for third-row transition elements,11 and other distances match
more closely. The trends are well reproduced, and in particular
there is a prediction of progressive weakening to the level of
reversibility: the duly observed reversibilities of the additions of
7514
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
Figure 6. Absorption spectra in CH2Cl2 of Pt2 1 (a) and corresponding
adducts Pt2-O2 3 (b), Pt2-SO2 5 (c), and Pt2-CO 4 (d).
Figure 5. 11B-{1H(broadband)} NMR spectra of aliquots of a single
solution of Pd2 8 in CD2Cl2/CH2Cl2 before (trace a) and after (trace b)
addition of CO to give Pd2-CO 11. Addition of SO2 to (b) quantitatively
affords Pd2-SO2 13 (trace c), but bubbling CO through the warmed
solution of 13 displaces the SO2 to reform Pd2-CO 11, (trace d). The
large “hump” in the spectra centered at δ(11B) about ꢀ5 ppm is due to
boron in the Pyrex NMR sample tube.
CO and SO2 to the PtPd 7 and Pd2 8 compounds are described
and discussed below.
Thus, for the CO adducts, the calculated and measured CꢀO
distances are equal within experimental uncertainties, namely,
1.174(7) [1.168] in Pt2-CO 4, 1.153(4) [1.162] in PtPd-CO 10
(NB: a considerable disorder in the single-crystal used in the
X-ray study renders this measured value less reliable), and
1.158(3) [1.158] in Pd2-CO 11. Plots of the calculated metal
separations versus the measured and calculated infrared stretch-
ing frequencies show a linear progression (Figure 4). Although
there is a systematic overestimation of the stretching frequency of
about 60 cmꢀ1, this mirrors the systematic errors in the distances
to the metals, and is consistent along the series.
In contrast to the diplatinum system, the addition of CO to
Pd2 8 is readily reversible, and removal of the solvent from
solutions of Pd2-CO 11 on a rotary evaporator results in the loss
of the CO to reform Pd2 8. CO may also be displaced by addition
of SO2 to form Pd2-SO2 13, but the SO2 is also not strongly
bound, and it may in turn be replaced by bubbling CO through
solutions of Pd2-SO2 13 in warm dichloromethane, as illustrated
in Figure 5. The sulfur dioxide unit may be similarly displaced
by CO from the mixed-metal adduct PtPd-SO2 12, although
it requires an extended time at the reflux temperature of CH2Cl2
to accomplish this. This reluctance is in line with the slightly
stronger binding to the PtPd vector suggested by the calculated
and measured intramolecular dimensions. In solution these
compounds probably form an equilibrium mixture involving
the complexed and uncomplexed metallaborane which is driven
fully to product by the presence of excess CO or SO2. A more
direct example showing presence of such an equilibrium is in
the SO2 adduct of [(phenanthroline)Pd(PMe2Ph)2PtB10H10],
(phen)PdPt 14, namely, [(phenanthroline)Pd(PMe2Ph)2Pt-
(SO2)B10H10], (phen)PdPt-SO2 15. Here, the 11B spectrum of
Figure 7. (A) Photorelease of CO after excitation of Pt2-CO 4 (ca. 100
μM in CH2Cl2) by a 425 nm laser pulse (ca. 2 mJ/pulse), measured at
524 nm. (B) Inset: transient absorption spectrum recorded 1 μs after
excitation by a 308 nm laser pulse (ca. 15 mJ/pulse) compared with the
normalized absorption spectrum of Pt2 1 (solid line) in CH2Cl2.
crystalline samples of (phen)PdPt-SO2 15 dissolved in CD2Cl2
shows that it forms an equilibrium mixture of 14 and 15 which
requires the addition of excess SO2 to push it to pure solutions
of 15.12
That O2, CO, and SO2 are released by the displacement or by
warming solutions differently according to the identity of the
metals, indicates that the reversibility of the binding is a tunable
feature which might be of interest for, inter alia, the design of
sensor materials and as handy sources of gas fluxes in a defined
environment. For many possible applications, an external trigger
to engender the instant release of the bound molecule would be
required. This led us to examine the photophysical properties of
the adducts, as photolysis by laser pulses using specific wave-
lengths could be a suitable and well-defined trigger for such a
process.
Absorption Spectra. Figure 6 shows absorption spectra of Pt2
1 and its O2, CO, and SO2 adducts. Compound Pt2 1 exhibits a
quite distinctive absorption band at 527 nm, which is well
separated from other electronic transitions. Reaction of Pt2 1
with O2, CO, or SO2 is instantaneous, and the adduct formation
is accompanied by the disappearance of this absorption band. In
the case of Pt2-SO2 5, absorption below 350 nm is obscured by
that of free SO2 (Figure S2c, Supporting Information). We have
included (phen)PdPt 14 and (phen)PdPt-SO2 1512 to provide
extra examples of the mixed metal species. The absorption
7515
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
changes observed for PtPd 7 and (phen)PdPt 14 are comparable
to those in Pt2 1, with the respective absorption bands at 495 and
528 nm disappearing after the addition of SO2 (Figures S3,
Supporting Information). Large absorption changes are also ob-
served after the formation of Pd2-SO2 13 from Pd2 8 (Figure S2,
Supporting Information). In all cases, the adducts have suffi-
ciently different absorption spectra when compared with the
original cluster compounds to allow the spectroscopic monitor-
ing of the reactions occurring in solution.
Nanosecond Laser Flash Photolysis. Nanosecond transient
absorption spectroscopy of the Pt2-CO 4 adduct shows the
absorption changes (Figure 7) that arise from the liberation of
CO from the bimetallic binding site. The changes occur within an
excitation pulse width (ca. 28 ns) (Figure 7A), and the transient
absorption spectrum nicely matches the absorption spectrum of
Pt2 1 (Figure 7B), confirming that the photoproduct is indeed
Pt2 1. Furthermore, it is apparent that CO molecules are liberated
from the bimetallic binding site very quickly. The diffusion-
controlled back-reaction proceeds on a millisecond time scale
(Figure S4, Supporting Information) that depends on the con-
centration of dissolved CO.
The quantum yield of the CO release, measured to be 0.60 (
0.03 upon 308 nm irradiation, shows a high efficiency of scission
of the Pt2-CO linkage. To the best of our knowledge, the
thermodynamics and the mechanism of CO photorelease and
rebinding have mostly been studied with heme model com-
pounds to understand processes through which heme proteins
perform their physiological function.13 However, these processes
are complex because of a heme center spin transition and, in
some cases, because of the formation of a four-coordinate heme
center or other factors, such as the flexing shape of the protein
crevice, and cannot therefore be compared to the behavior of the
Pt2-CO 4 adduct.
The results further show that Pt2-SO2 5 behaves similarly to
Pt2-CO 4 even though the absorption changes are smaller (data
not shown). Because of the absorption of dissolved SO2 below
350 nm, the quantum yield of the SO2 photorelease from Pt2-
SO2 5 could not be measured. The adducts Pd2-SO2 13, PtPd-
SO2 12, and (phen)PdPt-SO2 15 also instantly release SO2
(Figure S5, Supporting Information), and the liberated SO2
recombines with the corresponding cluster to give the original
adducts. There is little information in the literature on the
reversible binding of SO2 to metal complexes. For example,
although it has previously been shown that some four-coordinate
iridium species such as Vaska’s compound [IrCl(CO)(PPh3)2],2b
and platinum compounds containing the terdentate-coordinating
monoanionic [2,6-(CH2NMe2)2-C6H3]ꢀ ligand,3b,14 can reversi-
bly bind SO2, there are no reports, to our knowledge, on the release
of SO2 upon UVꢀvis irradiation.
Figure 8. Kinetics following about 130 fs UV-excitation of the O2, CO,
and SO2 adducts of (A) Pt2 1 and (B) Pd2 8 saturated with the added
respective gases. Excitation is at 395 nm, with a probing wavelength of
535 nm for Pt2-O2 3, Pt-CO 4, and Pt2-SO2 5, and of 540 nm for Pd2-CO
11 and Pd2-SO2 13. All kinetics are normalized to maxima, with the solid
lines corresponding to fits obtained from global fitting.
have measured its photoluminescence at 1270 nm (Figure S6,
Supporting Information), as emission at this wavelength arises
from relaxation of O2(1Δg) back to the triplet ground state
O2(3Σgꢀ). The quantum yield of O2(1Δg) formation, ΦΔ,
estimated using a porphyrin standard,16 is below 0.01, thus
indicating that the splitting channel leading to O2(1Δg) is of
minor importance, although the option of modifying the system
specifically to generate 1Δg dioxygen is a future possibility that we
are currently pursuing. Following photorelease, the bare diplati-
num unit Pt2 1 rebinds O2, albeit with some minor photode-
gradation being evident. Indeed, the lifetime of O2(1Δg), which is
a measure of its reactivity with dissolved molecules, is dictated by
the quenching efficiency of the metallaborane unit (Figure S6,
Supporting Information). As a result, the measured lifetime for
O2(1Δg) of 45 to 55 μs is much less that the reported value of
100 μs in pure CH2Cl2 in the absence of any singlet oxygen
quenching activity.16
For comparison purposes, it may be noted that some binuclear
μ-superoxo- and μ -peroxo cobalt complexes similarly produce
ground-state dioxygen upon irradiation.17 Some mono-nuclear
complexes also eliminate triplet-state dioxygen. For example,
[IrCl(O2)(CO)(PPh3)2] does so with a quantum yield of
up to 0.43 and, as with Pt2-O2 3, the quantum yield ΦΔ is low
at 0.03.18 Conversely, the photolysis of the oxodiperoxo-
molybdenum(VI) complex, [MoO(O2)2], produces O2(1Δg) with
ΦΔ up to 0.42 depending on the excitation wavelength.19
The complex producing O2(1Δg) upon photolysis, with a
behavior most similar to Pt2-O2 3, is the bis(triphenylphos-
phine)dioxygenplatinum peroxo complex [(Ph3P)2Pt(O2)]; for
this compound, this behavior is attributed to a ligand (i.e., O2)
transition with a ligand-to-metal (O2 f Pt) charge transfer
contribution.20 Unfortunately, no ΦΔ values were given. To
the best of our knowledge, Pt2-O2 3, is the only example of a
Of the bimetallic boron clusters reported here, only Pt2 1 binds
dioxygen.1b Like Pt2-CO 4, the photorelease of O2 from Pt2-O2 3
occurs with a high quantum yield of 0.58 ( 0.03 after excitation
by a 308 nm pulse. In addition to dioxygen in the ground triplet
state [i.e., O2(3Σgꢀ)], a contribution of singlet oxygen O2(1Δg),
was detected. This is surprising, as O2(1Δg) is most often
produced by energy transfer to the dioxygen ground state from
an electronically excited triplet state of a photosensitizer
molecule.15 Under our conditions, we can exclude the photo-
sensitized formation of O2(1Δg) as no long-lived excited states
were detected (see below). Thus, the O2(1Δg) formation is
clearly a result of the relaxation of excited states of the dioxygen
adduct. To assess the efficiency of the O2(1Δg) generation we
7516
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
Figure 10. Transient absorption spectra of (A) Pt2-SO2 5, Pt2-CO 4,
Pt2-O2 3, and (B) Pd2-SO2 13, Pd2-CO 11 adducts recorded at 200 ps
after excitation at 395 nm. Difference absorption spectra given as a
difference between absorption spectra of the Pt2 1 and Pd2 8 clusters and
the corresponding adducts are shown as solid lines for comparison.
Figure 9. Transient absorption spectra of (A) Pt2-SO2 5, (B) Pt2-CO 4,
and (C) Pt2-O2 3 measured after two different delay intervals, 3 ps
(black) and 100 ps (red), after excitation at 395 nm.
(Figure 9). The following rise, occurring with an 11 ps time
constant in Pt2-CO 4 and 20 ps in Pt2-O2 3, describes the
formation of Pt2 1 in the ground state after the gas photorelease
(Figures 9 and 10). In contrast, the excited-state relaxation of Pt2-
SO2 5 can be satisfactorily described by at least three decay
components of 1, 6, and 80 ps. The photoproduct generated
within the time window of the experiment is not ground-state Pt2
1 (Figure 10). Instead, an intermediate complex, which may be,
for example, an excited Pt2-SO2 with one SꢀPt bond broken, is
formed within the first 200 ps, and the complete release of SO2
from the Pt2-SO2 adduct probably takes place on a longer time
scale inaccessible in our femtosecond experiments (>1 ns). This
indicates that the binding affinity of SO2 toward the Pt2 1 cluster
is markedly greater than that of O2 or CO, consistent with the
conclusions from the results of DFT calculations (Supporting
Information) and of experimental displacement reactions dis-
cussed in the section above.
For the dipalladium species, the kinetics recorded after the
excitation of the Pd2-based adducts shown in Figure 8 demon-
strate a different binding to the Pd2 8 and Pt2 1 compounds for
SO2 compared to CO. Thus, the adduct Pd2-SO2 13 behaves
comparably to Pt2-SO2 5 and the excited states decay multi-
exponentially with time constants of 0.3, 2.5, and 40 ps. This
behavior demonstrates that the nature of the metal binding sites
quantitativelyaffects therelaxationofthewhole complex andthatit
isfasterin Pt2-SO2 5. In contrast, however, the behavior of Pd2-CO
11 qualitatively differs from that of Pt2-CO 4, as shown by the
features of the transient spectra at 200 ps, which differ signifi-
cantly from the difference absorption spectrum of Pd2/Pd2-CO
(Figure10). While excitationofPt2-CO 4leads tothephotorelease
of CO with an 11 ps time constant, Pd2-CO 11 undergoes excited-
state relaxation with no formation of ground state Pd2 8 within the
measured time window. For comparison here, it may be noted that
the ultrafast kinetics of CO photorelease have been studied in
bimetallic complex that reversibly binds dioxygen and releases it
upon irradiation.
Femtosecond Transient Absorption. To obtain further in-
sight into the photorelease mechanism of CO, SO2, and O2
molecules from the adducts following UVꢀvis irradiation, we
studied the ultrafast dynamics upon excitation with pulses of
about 130 fs duration centered at 395 nm. The resulting kinetics
for the Pt2- and Pd2-based homometal adducts measured at 535
and 540 nm (the wavelengths where the photoproduced ground-
state Pt2 1 and Pd2 8 compounds absorb) are shown in Figure 8.
The kinetics of both of the compounds Pt2-CO 4 and Pt2-O2 3
reveal an initial fast decay followed by a rise component, whereas
no corresponding rise component is observed for Pt2-SO2 5. To
identify the origin of this qualitatively different behavior, we have
compared the transient absorption spectra recorded after differ-
ent delay intervals. While the transient absorption spectra of Pt2-
SO2 5 do not significantly differ in shape when recorded after
delays of 3 and 100 ps, the broad and featureless spectra of Pt2-O2
3 and Pt2-CO 4 after 3 ps are replaced by a well-defined band
centered at about 535 nm after a 100 ps delay (Figure 9). Further,
a comparison of the transient absorption spectra at 200 ps with
the difference absorption spectra (Figure 10) shows that,
whereas the 200 ps spectra for Pt2-O2 3 and Pt2-CO 4 match
well the absorption spectrum of Pt2 1 in the ground state, the
transient absorption spectrum of Pt2-SO2 5 peaks at 490 nm, and
is thus significantly different from the 535 nm maximum.
The initial signals generated from Pt2-CO 4 and Pt2-O2 3
decay with a about 1.5 ps time constant. Since the transient
absorption spectra at 3 ps do not contain bands attributable to
the compound Pt2 1, this decay can be assigned to a fast
relaxation of the excited states of the corresponding adducts
7517
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
Figure 11. (A) Kinetics of the SO2 complexes after excitation at
395 nm. Probing wavelengths were 530 nm for Pt2-SO2 5, 540 nm for
Pd2-SO2 13, 520 nm for (phen)PdPt-SO2 15, and 505 nm for PtPd-SO2
12. The bottom panel (B) shows an expansion of the first 10 ps of the
traces in the upper panel.
Figure 12. (A) Release and rebinding of O2 under cyclic irradiation of
Pt2-O2 3 in CH2Cl2 (140 μM solution). The absorbance changes
recorded at 524 nm during irradiation (light ON, red line) and dark
periods (light OFF, black line), respectively, indicate the release of, and
reverse binding of, O2. The oxygen-saturated solution was irradiated by a
Xe lamp (>370 nm). (B) Continuous irradiation for about 100 s under
the same conditions as above.
detail in some heme compounds.21,22 In these, the CO ligands are
photodissociated instantaneously, and vibrational cooling of the
ferrous ground state takes place in a few ps. Geminate recombina-
tion occurs from the nanosecond to millisecond time scales, and
the CO molecules are stored within the protein matrix in dock-
ing sites.
clusters with mixed metal sites the photoinduced release of SO2 is
significantly faster than for the SO2 adducts 5 and 13 of the
homometal compounds Pt2 1 and Pd2 8, most likely because of
the metalꢀmetal bonding asymmetry that is present in the mixed
metal clusters, and which would reinforce the sequential asym-
metric scission of the M-SO2 bonds suggested above for the
homometal compounds.
Since both homometal SO2 adducts Pt2-SO2 5 and Pd2-SO2
13 adducts have a comparable behavior, we also studied ultrafast
dynamics of two SO2-based adducts with mixed-metal binding
sites, namely, PtPd-SO2 12 and (phen)PdPt-SO2 15 (Figure 11).
The results clearly show that the nature of the metal atoms alters
the amplitude and time constant of the initial fast component.
In addition, the ultrafast dynamics are more complicated in
(phen)PdPt-SO2 15, showing a 0.8 ps rise that may be assigned
to the formation of (phen)PdPt 14 because of the subpicosecond
release of SO2. This conclusion is also supported by a reasonable
match between the transient absorption spectrum of (phen)-
PdPt-SO2 15 taken at 200 ps and the corresponding difference
absorption spectrum (Figure S7, Supporting Information). Thus,
(phen)PdPt-SO2 15 may actually release the SO2 molecule on a
subpicosecond time scale, and the further dynamics, which are
comparable to other SO2-bound adducts, may then reflect a
relaxation of (phen)PdPt 14 that is generated in a “hot” state.
The time constants of this relaxation, as obtained from global
fitting, are 2.5 and 35 ps. A similar situation occurs for PtPd-SO2
12, for which the transient absorption spectrum recorded at
200 ps matches the PtPd/PtPd-SO2 difference absorption spec-
trum (Figure S7, Supporting Information). However, no rise is
detected; suggesting that for this compound the SO2 release is
ultrafast and takes place within the excitation pulse. In accord
with the hypothesis of ultrafast SO2 release in PtPd-SO2 12,
comparison of the transient absorption spectra measured at 0.25
and 200 ps (Figure S8, Supporting Information) shows that
already at 250 fs the transient absorption spectrum matches
the difference absorption spectrum. Thus, we conclude that in
Continuous Irradiation. The reversible uptake of the studied
gas molecules under UVꢀvis irradiation described above also
may be observed on the macroscopic scale, and we also probed
the reversibility of the adduct formation, as well as compound
photostability, under the condition of continuous irradiation (λ >
370 nm). With Pt2-O2 3, the cycling of a sequence of irradiation
periods of 20 s, alternating with dark intervals, shows that nearly
60% conversion to the dioxygen-free compound Pt2 1 is reached
in about 10 s of irradiation (Figure 12A). The rebinding of O2
occurs within 20 s in the dark interval. The photostability of the
Pt2 1 compound under the irradiation conditions used is demon-
strated by longer irradiation periods which lead to a constant
stationary concentration, also about 60% (Figure 12B). It there-
fore appears that the release and rebinding of O2 can be cycled
with high efficiency without considerable photodegradation of
the Pt2 cluster. Pt2-CO 4 behaves similarly (Figures S9, S10,
Supporting Information), but, in agreement with the higher
affinity of CO toward Pt2 1 compared to O2, the maximum
stationary concentration of photoproduced 1 is about 10%,
rather than the 60% observed for Pt2-O2 3. The Pt2-SO2 5
complex, in which the SO2 molecule is the most strongly bound
of the series reported here, does not show release under these
7518
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
Table 2. Crystallographic Data for Pd2 8, PtPd-CO 10, Pd2-CO 11, PtPd-SO2 12, and Pd2-SO2 13
a
8
10
11
12
13
formula
difractometer
size
C
32H54B10P4Pd2
C33H54B10OP4PdPt C33H54B10OP4Pd2 CHCl3 C32H54B10O2P4PtPdS CH2Cl2 C32H54B10O2P4Pd2S C6H6
3
3
3
Nonius KappaCCD Br€uker X8 Apex
Br€uker X8 Apex
Nonius KappaCCD
0.07 ꢁ 0.06 ꢁ 0.01
1064.76
Nonius KappaCCD
0.31 ꢁ 0.18 ꢁ 0.13
1025.70
0.16 ꢁ 0.15 ꢁ 0.12 0.35 ꢁ 0.28 ꢁ 0.03 0.19 ꢁ 0.12 ꢁ 0.02
883.53
formula wt.
color, shape
crystal sys.
space group
a/Å
1000.23
1030.97
orange prism
monoclinic
C2/c
orange plate
orthorhombic
Pna21
red prism
monoclinic
P21/c
yellow plate
orthorhombic
Pna21
red fragment
orthorhombic
Pc21n
23.4332(3)
11.4378(2)
17.2521(3)
122.771(1)
3888.03(11)
1.509
23.566(5)
10.827(2)
17.533(4)
90
12.4922(11)
24.4954(22)
15.7457(13)
97.903(4)
4772.4(7)
1.435
23.901(5)
10.743(2)
17.573(4)
90
23.964(5)
10.788(2)
17.682(4)
90
b/Å
c/Å
β/deg
V/Å3
4473.6(16)
1.485
4512.2(16)
1.65
4571.4(16)
1.49
Dcalc/g cmꢀ3
3
F000
1792
1976
2080
2216
2088
data/restr/par
extinction parameter
μ mmꢀ1
4457/0/221
11046/1/548
11803/0/507
9819/1/502
0.00114(10)
3.833
8929/10/498
1.114
3.694
1.083
1.006
GOF
1.055
1.032
1.023
1.025
1.034
R1 (I > 2σ(I))
wR2 (all data)
0.0323
0.0761
0.0243
0.0361
0.0469
0.0433
0.0631
0.0879
0.0961
0.1038
F
max and min /eꢀ Åꢀ3 0.83 and ꢀ1.42
0.66 and ꢀ1.46
0.022(5)
1.246 and ꢀ1.162
1.11 and ꢀ1.05
0.026(8)
1.25 and ꢀ0.97
Flack
a Common factors. All measured at 150(2) K, Z = 4.
conditions whereas PtPd-SO2 12, which binds SO2 less strongly,
exhibits well-measurable changes because of the release of SO2
(Figure S11, Supporting Information). Pd2-SO2 13 also releases
SO2 under these conditions, but now the release and rebinding
sequence is accompanied by significant compound photodegradation.
are all addressed in this work. In this context it is noteworthy that
the dioxygen adduct Pt2-O2 3 releases dioxygen upon UVꢀvis
irradiation with high quantum yields, and, moreover, a fraction of
the dioxygen is released in singlet configuration.
The results show that the tailoring of this system by variation
of metal is highly viable and useful, and that DFT predictions are
a very useful aid to predicting viable targets and therefore
optimizing the use of valuable synthetic bench time. The results
augur well for future tailoring and fine-tuning of this system by
variation of metal and metal ligand, by variation of boron cluster
substituent, or by cluster constituents, such as replacing B by C,
S, and so forth (as a fourth way of modification), and also for
extension to other small molecules, not only in terms of
sequestration, but also in terms of activation. In this latter regard
there is potential here for these B-frame bimetallics to rival the very
extensively investigated area of so-called “A-frame” bimetallics in
versatility and consequent utility, and we hope to explore all these
possibilities.
’ CONCLUSIONS
Predictive DFT work suggested that PtPd 7 and Pd2 8 would
be synthetically viable targets, and that CO and SO2 would bind
to them in an analogous manner to that established for Pt2 1. In
addition, DFT calculations suggested that the Pd2 8 would bind
SO2 and CO more weakly than the corresponding diplatinum
compound Pt2 1, thus allowing for reversibility observed earlier
for the uptake of SO2 and CO addition in a similar manner to that
observed previously for the reversible uptake of O2 by Pt2 1.
Experiments then successfully showed that the binding of the
molecules is indeed progressively weaker in the sequence {PtPt} f
{PtPd} f {PdPd}. The binding of CO is shown to be weakened
sufficiently that ready reversibility of CO and SO2 uptake was
observed for the Pd2 8 bimetallic site and to a lesser extent for the
PtPd 7 compound.
We found that the sequestered molecules could be ejected by
UVꢀvis irradiation. Very interesting consequences follow from
the discovery of the light-activated control of the gas molecule
release. Fast and spatiotemporal control of active molecule
delivery are fundamental criteria for any potential application of
these gas-releasing systems,23 and the on/off triggering by light
described here satisfies both of these requirements. Other addi-
tional important features imposed on such delivery systems are
photostability, the ability to reversibly bind the active molecules,
the possibility to follow the course of the binding/release pro-
cesses, and the accurate prediction of system improvements; these
’ EXPERIMENTAL SECTION
General Procedures. All reactions were carried out under stan-
dard vacuum or inert-atmosphere techniques, although subsequent
operations were generally carried out in air, except for a number of
those involving solutions of the oxygen-sensitive [(PMe2Ph)4PtPd-
B10H10] 7 and [(PMe2Ph)4Pd2B10H10] 8. B10H14 was obtained com-
mercially (Katchem, Prague). Dichloromethane, hexane, and pentane
(Fluka) were dried over appropriate desiccants and stored in 500 mL RB
flasks fitted with Teflon vacuum taps and transferred by vacuum
distillation into vessels cooled by liquid N2. KHBEt3 (Aldrich, 1.0 M
THF solution) was used as received. Other starting materials were of
reagent or analytical grade and were used as purchased. Reactions were
generally carried out in a 5 mm o.d. NMR tube with a vacuum tight
7519
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
Teflon screw cap. [cis-PtCl2(PMe2Ph)2] was made by the literature
route24 from stirred water mixtures of K2[PtCl4] and PMe2Ph.
[(PMe2Ph)2PtB10H12] was made essentially according to the method
reported earlier25 in yields of about 90%. [PdCl2(PMe2Ph)2] was
prepared using ethanol solutions of K2[PdCl4] and PMe2Ph from which
the [PdCl2(PMe2Ph)2] precipitates. The resultant yellow suspension
was filtered and washed with a small amount of alcohol to obtain
the [PdCl2(PMe2Ph)2] and was used without further purification.
[(PMe2Ph)2PdB10H12] was made by addition of [PdCl2(PMe2Ph)2]
to a stirred solution of B10H14 in CH2Cl2 that also contained 2 equiv of
1,8-bis-(dimethylamino)naphthalene (“Proton Sponge”), followed by
flash chromatography down a silica-gel column (70/30 CH2Cl2/
n-hexane). To achieve a maximum yield of [(PMe2Ph)2PdB10H12] the
ethanol filtrate from the preparation of [PdCl2(PMe2Ph)2] was reduced
to dryness on a rotary evaporator, redissolved in CH2Cl2 and added to a
CH2Cl2 solution containing excess B10H14 and Proton Sponge; the
mixture was then stirred overnight and then also subjected to flash
column chromatography. This procedure allows an almost quantitative
overall yield of [(PMe2Ph)2PdB10H12].
(100 mL), and the solution heated at reflux under nitrogen overnight.
The color slowly changed from yellow to brown/dark purple. After
cooling and rapid filtration in air, washing with benzene gave micro-
crystalline Pt2 1 (0.157 g, 0.148 mmol; 53% yield based on Pt).
In both cases A and B, purification of Pt2 1 is better done by filtration
of CH2Cl2 solutions in air and crystallization from aerated CH2Cl2/
hexane solution stored at < ꢀ20 ꢀC. This procedure gives the orange-
colored Pt2-O2 3, from which compound Pt2 1 may be easily regenerated
in situ for further experimentation by purging solutions of the com-
pound with an inert gas.
Preparation of [(PMe2Ph)4PtPdB10H10] (PtPd 7). The above proce-
dure for 1was followed with [(PMe2Ph)2PtB10H12] (0.099 g, 0.167 mmol)
in toluene (23 mL) and KHBEt3 (0.34 mL, 0.34 mmol) added. After
allowing to warm to ꢀ30 ꢀC the temperature was reduced to ꢀ50 ꢀC
and [PdCl2(PMe2Ph)2] (0.076 g, 0.168 mmol) added against a flow of
nitrogen, and the reaction mixture allowed to warm slowly to room
temperature with stirring. Subsequently, a purple-brown solid was
obtained by filtration after washing with benzene and pentane. The
solid was dissolved in CH2Cl2, passed through a sintered glass filter,
crystallized from CH2Cl2/n-pentane at about ꢀ20 ꢀC, and identified
it as [(PMe2Ph)4PtPdB10H10] by multielement NMR spectroscopy
(51 mg, 0.052 mmol, 31%). Elem. anal. for [(PMe2Ph)4PtPdB10H10].
CH2Cl2, found(calc.): C37.57(37.49) H 5.14(5.34). Attempts to
obtain single-crystals suitable for X-ray analysis from dichloromethane
hexane or chloroform hexane solutions were unsuccessful. NMR data
(298 K, CD2Cl2), δ(11B) [1H]/ppm: BH(7,11), +31.9 [+5.23]; BH-
(9,3,6,4,5,12) +25.3 [+4.67, +4.51(2), +4.35(2), +3.91]; BH(8,10),
1
Criteria of purity consisted of clean 11B, H, and 31P NMR spectra
consistent together with molecular and crystal structures obtained by
single-crystal X-ray diffraction analyses. A range of mass spectrometry
techniques were attempted on all the compounds reported here but
none gave identifiable fragments.
NMR spectroscopy was performed at about 5.9 and 11.75 T (fields
corresponding to nominal 250 and 500 MHz 1H frequencies respectively)
using commercially available instrumentation and using techniques and
procedures as adequately described and enunciated elsewhere.26 Chemical
shifts δ are given in ppm relative to Ξ = 100 MHz for δ(1H) ((0.05 ppm)
(nominally tetramethylsilane, Si(CH3)4), Ξ = 32.083972 MHz for δ(11B)
((0.5 ppm) (nominally Et2OBF3 in CDCl3) and Ξ = 40.480730 MHz for
δ(31P) ((0.5 ppm) (nominally 85% aqueous H3PO4), calibrations being
effected by use of solvent deuteron or residual proton signals as internal
secondary standards. Ξ is as defined by McFarlane.27 Cluster 11B and 1H
NMR data are presented as assignment δ(11B) [δ(1H) for directly bound
H(exo) in brackets]. IR spectra were recorded on a Perkin-Elmer
Spectrum One FT-IR spectrometer using a diamond cell or on a KBr
plate in a Nicolet Nexus 670 FT-IR spectrometer.
1
ꢀ23.0 [+1.14]. δ(1H) PMe2(Pt)A: +1.71 {N(31Pꢀ H) 9.8 Hz, 3J-
1
1
3
(
(
195Ptꢀ H) 24.6 Hz)}, PMe2(Pt)B: +1.62 {N(31Pꢀ H) 9.2 Hz, J-
1
1
195Ptꢀ H) 21.4 Hz)}, N(31Pꢀ H) ca. 10 Hz; PMe2(Pd)A: +1.47
{N(31Pꢀ H) 8.2 Hz}, +1.45, N(31Pꢀ H) 7.7 Hz}, and for PPh +7.17 to
1
1
+7.41 {multiplets}. δ(31P)/ppm, CD2Cl2, ꢀ50 ꢀC: on Pt +3.5 [3J(31Pꢀ
31P) 10, 1J(195Ptꢀ P) 2640 Hz ], on Pd ꢀ21.1 [3J(31Pꢀ P) 10 Hz,
31
31
possible 2J(195Ptꢀ P) ca. 40 Hz].
31
Preparation of [(PMe2Ph)4Pd2B10H10] (Pd2 8). The above procedure
as described for Pt2 1 was followed using toluene (100 mL) and
[(PMe2Ph)2PdB10H12] (0.203 g, 0.403 mmol), with 0.81 mL, 0.81 mmol,
of KHBEt3 solution injected at ꢀ70 ꢀC. The mixture was allowed to warm
slowly with stirring to ꢀ50 ꢀC and held at that temperature for 30 min
during which time the mixture turned brown. [PdCl2(PMe2Ph)2] (0.443 g,
0.403 mmol) was added against a flow of nitrogen, and the reaction mixture
left to warm slowly to room temperature with stirring overnight. Next
morning the mixture was quickly filtered and washed with benzene giving
0.14 g of a purple-brown solid. A further 0.065 g of solid was scraped from
the walls of the reaction vessel. Both solids were identified as [(PMe2Ph)4-
Pd2B10H10] (compound 8, combined yield 0.234 g, 0.23 mmol; 57%).
Dark red single-crystals were obtained from saturated CDCl3 solutions of
the compound under anaerobic conditions stored at ꢀ20 ꢀC. Elem. anal.
C32 H54 B10 P4 Pd2, found(calc.): C 43.55(43.48), H 6.17(5.97). NMR
data (298 K, CDCl3): BH(9,12) +29.7 [+4.41], BH(3,6) ca. +32.6 [+4.85]
and (near-coincident resonances) BH(4,5,7,11) ca. +32.6 [+5.48], and
BH(8,10) ꢀ20.1 [+0.99]; with δ(1H) for PMe2 +1.49 and +1.72 {both
Calculations were carried out using the Gaussian 03 package.10a
Structures were initially optimized using standard ab initio methods with
the STO-3G* basis-sets for B, C, P, S, N, and H, with the LANL2DZ
basis-set for the metal atoms. The final optimizations, including fre-
quency analysestoconfirm thetrueminima, were performed using B3LYP
methodology with 6-31G* and LANL2DZ basis-sets.10bꢀe In all cases the
compounds were modeled using hydrogen atoms rather than organyl
groups on the phosphine ligands to reduce computation time. Computed
carbonyl frequencies were scaled with a 0.9613 scaling factor.28
Syntheses. Improved Preparation of [(PMe2Ph)4Pt2B10H10] (Pt2 1).
(Case A) Toluene (20 mL) was injected through a septum into a two-
neck round-bottomed flask containing [(PMe2Ph)2PtB10H12] (0.202 g,
0.341 mmol) and a stir bar. The mixture was cooled in a dry ice bath to
ꢀ71 ꢀC and KHBEt3 (0.68 mL, 0.68 mmol) then injected, after which the
mixture was allowed to warm slowly to ꢀ30 ꢀC, giving a colorless solution.
The temperature was lowered to ꢀ50 ꢀC, [PtCl2(PMe2Ph)2] (0.155 g,
0.34 mmol) added against a flow of nitrogen, the stirring continued,
and the mixture allowed to warm slowly to room temperature. A clear
orange solution quickly formed which gradually darkened with a dark
solid appearing on the bottom of the flask. The next day the mixture
was filtered giving 0.178 g of dark purple microcrystalline solid which
was quickly washed with benzene, then pentane. A further 0.096 g was
scraped from the wall of the reaction vessel affording 0.274 g,
0.258 mmol, 76% of Pt2 1.1b
1
with N(31Pꢀ H) ca. 8 Hz}, and for PPh +7.22, +7.34 and +7.40 4H
{multiplets}; also δ(31P) ꢀ15.3 ppm.
Reaction of Pd2 8 with CO to Give Pd2-CO 11. CD2Cl2 (ca. 0.5 mL)
was condensed into an NMR tube containing a sample of Pd2 8 (ca. 3
mg); the tube was then sealed, removed from the Schlenk line, and
shaken to dissolve the sample. The tube was reattached to the Schlenk
line and opened to an atmosphere of pure CO. A pale tan color was
observed to diffuse down slowly from the top of the dark purple-red
solution. Agitation of the tube eventually resulted in the disappearance
of the purple-red color to form a tan-colored solution, and 11B, 31P, and
1H-{11B} NMR spectroscopy then indicated a complete conversion to
Pd2-CO 11. Removal of the solvent from CH2Cl2 solutions of the
(Case B) PMe2Ph (0.2 mL, 1.4 mmol) was injected into a suspension
of [(PMe2Ph)2PtB10H12] (0.342 g, 0.557 mmol) in degassed toluene
7520
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
1
compound on a rotary film evaporator (water bath, ca. 30 ꢀC) resulted in
regeneration of Pd2 8, pure by NMR spectroscopy. Crystals suitable for
single-crystal X-ray work were obtained by diffusion of n-pentane
through a layer of benzene into a CHCl3 solution of the compound.
Because of the easy loss of CO from the compound, the sample was
agitated slightly to accelerate mixing and to initiate crystallization, but
this caused the formation only of small crystals, which necessitated the
use of a rotating-anode X-ray source to obtain a suitable data set for the
diffraction analysis. Infrared, diamond cell: ν(BH) 2511br, ν(CO)
1882 cmꢀ1. NMR data (298 K, CDCl3): BH(9,12) ca. +28.5 [+4.64]
and (coincident boron resonances) BH(3,6) ca. +28.5 [+2.16]}, BH-
(4,5,7,11) +19.6 [+4.33] and BH(8,10) ꢀ11.3 [+1.98]; with δ (1H) for
The [1Hꢀ H] correlation behavior of the individual doublet compo-
nents suggest that this fine structure probably arises from phosphorus-
to-proton couplings.
Single-Crystal X-ray Diffraction Analyses. Diffraction data
were collected either on a Nonius KappaCCD diffractometer with a
conventional sealed-tube X-ray source (λ = 0.71073 Å), or, for very small
crystals, on a Br€uker X8 Kappa APEX II CCD area-detector diffract-
ometer with an FR591 rotating anode generator. In each case crystals
were cooled with a dinitrogen gas stream. Absorption corrections were
made by semiempirical methods. The structures were solved by auto-
matic direct methods and refined by least-squares methods on all
measured F2 values, with a weighting scheme wꢀ1 = σ2(Fo 2) + (aP)2 +
(bP), where P = (Fo2 + 2Fc2)/3. Residuals were defined by R1 = ∑||Fo| ꢀ
1
PMe2 +1.26 and +1.53 {both with N(31Pꢀ H) ca. 10 Hz}, and for PPh
√
|Fc||/∑|Fo|, wR2 = [∑w(Fo ꢀ Fc2)2/∑w(Fo ) ]. Methods and pro-
grams were standard.29 Crystallographic data are summarized in Table 2.
The programs X-Seed30 and Ortep-3 were31 used as an interface to the
SHELX programs, and to prepare the figures. Compounds [(PMe2Ph)4-
(CO)PtPdB10H10], PtPd-CO 10, showed a disorder involving the
positions of the platinum and palladium atoms. A model of the disorder
was established assuming exchange in their positions and was refined
with a free variable. Further disorder involved two positions for one
phosphine group plus a slight twist in one phenyl group on a second
phosphine and these were also refined with a common occupancy and
refined with a free variable. Disordered dichloromethane molecules were
also apparent and were incorporated in the model using Platon/
Squeeze.32 [(PMe2Ph)4(SO2)PtPdB10H10] 12 also showed a similar
disorder but only involving an exchange in the positions of the platinum
and palladium atoms.
2
2 2
+7.05 to +7.40 {multiplets}]; also δ(31P) ꢀ14.1 ppm.
Reaction of Pd2 8 with SO2 to Give Pd2-SO2 13. The procedure
followed that described above for Pd2-CO 11. A golden coloration,
rather than the tan of the CO reaction, diffused down the column of
solution in the tube. Quantitative conversion was confirmed by NMR
spectroscopy. Crystals suitable for single-crystal X-ray diffraction anal-
ysis were obtained by diffusion of n-pentane through a thin layer of
benzene into a CH2Cl2 solution of the compound. NMR data (298 K,
CD2Cl2): BH(9,12) +31.4 [+4.58], BH(3,6) +34.7 [+2.65], BH-
(4,5,7,11) +21.3 [+4.23] and BH(8,10) ꢀ6.2 [+2.09]; with δ(1H) for
1
PMe2 +1.25 and +1.44 {both with N(31Pꢀ H) ca. 10 Hz}, and for PPh
+7.25 to +7.50 {multiplets}; also δ(31P) ꢀ8.35 ppm.
Reaction of PtPd 7 with CO to Give PtPd-CO 10. The procedure
carried out for compound 11 was repeated. Very dark red crystals
suitable for single-crystal X-ray diffraction analysis were obtained by
diffusion of n-pentane into CH2Cl2 solutions of the compound. NMR
data (298 K, CD2Cl2): BH(9) ca. +23.9 [+5.29], BH(12) ca. +20.7
[+4.18]; BH(3,6) +16.6 [+1.35], BH(4,5) +14.6 [+4.08], BH(7,11)
Photophysical and Photochemical Methods. Nanosecond
Transient Absorption. Laser flash photolysis experiments were per-
formed with a Lambda Physik COMPEX 102 excimer (308 nm, pulse
width 28 ns, incident energy up to 15 mJ/pulse) or a Lambda Physik
FL 3002 dye laser (425 nm, incident energy ∼1 mJ/pulse). Transient
absorption spectra were measured within 450ꢀ600 nm on a laser
kinetic spectrometer (Applied Photophysics, U.K.). The time profiles
were recorded using a 250 W Xe lamp equipped with a pulse unit and a
R928 photomultiplier (Hamamatsu). The quantum yields for the gas
release were determined by the comparative technique using benzo-
phenone as a standard (λex = 308 nm, ΦT = 1.0, Δε = 7220 Mꢀ1 cmꢀ1
at 530 nm.33
Time-Resolved Luminescence of Singlet Oxygen. The near-infrared
luminescence of O2(1Δg) at 1270 nm was monitored using a Ge detector
(1270 nm interference filter, Judson Ge diode) upon laser excitation by a
Lambda Physik COMPEX 102 excimer (λexc = 308 nm, fwhm 28 ns,
incident energy up to 200 mJ/pulse). Emitted luminescence was detected
at the right angle to excitation light. The signal from the detector was
collected in a 600 MHz oscilloscope (Agilent Infiniium) and transferred to
a computer for further analysis. The signal-to-noise ratio of the signals was
improved by the averaging of 100 to 500 individual traces. The short-lived
signal producedby the scattering of anexcitationlaserpulse was eliminated
by exciting the argon-saturated sample and by subtracting the obtained
signal from the signal recorded in oxygen-saturated solution. The quantum
yield of singlet oxygen formation, ΦΔ, was estimated by the comparative
method using 5,10,15,20-tetrakis(pentafluorophenyl)porphyrin as a stan-
dard (ΦΔ = 0.8).16
1
+7.3 [+3.80 {J(195Ptꢀ H) ca. 55 Hz}] and BH(8,10) ꢀ13.9 [+2.09];
1
1
with δ(1H) for PMe2(Pt)A +1.47 {N(31Pꢀ H) 10.0 Hz, 3J(195Ptꢀ H)
17.5 Hz)}, PMe2(Pt)B +1.41 {N(31Pꢀ H) 10.4 Hz, 3J(195Ptꢀ H) 22.5
1
1
Hz}, PMe2(Pd)A +1.35 {N(31Pꢀ H) 9.6 Hz}, PMe2(Pd)B, +1.28
1
{N(31Pꢀ H) 9.8 Hz} and for PPh +7.28 to +7.47 {multiplets}; also
1
δ(31P) ꢀ24.6 [1J(195Ptꢀ P) 2644 Hz ] and ꢀ18.1 [3J(195Ptꢀ P) 63
31
31
Hz]. Infrared, evaporated CH2Cl2 solution on to a KBr plate: ν(BH)
2511br, ν(CO) 1847 cmꢀ1
.
Reaction of PtPd 7 with SO2 to Give PtPd-SO2 12. The addition of
SO2 induced a ginger coloration. Green-yellow microcrystals were
obtained from n-pentane/CH2Cl2 solution, and yellow-plate single-
crystals for X-ray diffraction analysis were obtained by diffusion of n-
pentane into a CH2Cl2 solution of the compound. Anal. [(PMe2Ph)4
(SO2)PtPdB10H10].CH2Cl2, found(calc.): C 35.61(35.35), H 5.07-
(5.03). NMR data (298 K, CDCl3): BH(12) ca. +27.4 [+5.76
4
1
{antipodal coupling J(195Ptꢀ H) ca. 30 Hz}(see text in Supporting
Information, page S2)], BH(9) ca. +27.4 [+4.73]; BH(3,6) +24.3
[+2.15], BH(4,5) +18.0 [+4.36], BH(7,11) +9.8 [+4.08] and BH(8,10)
1
ꢀ7.6 [+2.96 {J(195Ptꢀ H) ca. 55 Hz}]; with δ(1H) for PMe2(Pt)A
1
1
+1.49 {N(31Pꢀ H) 10.0 Hz, 3J(195Ptꢀ H) 17.5 Hz)}, PMe2(Pt)B
1
1
+1.72 {N(31Pꢀ H) 10.4 Hz, 3J(195Ptꢀ H) 22.5 Hz}, PMe2(Pd)A
1
1
+1.37 {N(31Pꢀ H) 9.6 Hz}, PMe2(Pd)B, +1.57 {N(31Pꢀ H)
9.8 Hz} and for PPh +7.30 to +7.55 {multiplets}; also δ(31P) ꢀ10.7
[1J(195Ptꢀ P) 2605 Hz ] and ꢀ13.3 [no J(195Ptꢀ P) resolved].
31
3
31
[1Hꢀ H]-COSY interactions between 1H(4,5) at +4.36 and 1H-
Femtosecond Transient Absorption. Femtosecond pulses were ob-
tained from a 1 kHz femtosecond laser system (Integra-I, Quantronix).
The output pulses had ∼130 fs width, an average energy of ∼2 mJ/pulse,
and a central wavelength of 790 nm. The pulses were divided into two
paths serving as excitation and probe pulses. Excitation pulses centered
at 395 nm were generated by frequency doubling the primary output in a
BBO crystal, while a white-light continuum obtained by focusing a
fraction of the 790-nm pulses into a 0.2 cm sapphire plate was used for
probing. For signal detection, the probe beam and an identical reference
1
1
(PMe2)(Pd) at +1.37 and +1.57, and similarly between H(7,11) at
+4.08 and 1H(PMe2)(Pt) at +1.49 and +1.72, were observed, implying
small but finite longer-range couplings 5J(1HꢀBꢀPdꢀPꢀCꢀ H)-
1
(cisoid) and 5J(1HꢀBꢀPtꢀPꢀCꢀ H)(cisoid), but no correspondingly
1
observable transoid interactions. Additionally, in 1H-{11B} spectra, fine
doublet splittings on 1H(4,5) and 1H(7,11) of about 10 Hz were
observable. These were more apparent in the [1Hꢀ H]-COSY-{11B}
1
plots, which also showed a doublet splitting on 1H(8,10) of about 3 Hz.
7521
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
beam were focused onto the entrance slit of a spectrograph, which then
dispersed both beams onto a dual photodiode array detection system
(ExciPro, CDP Systems). A 1-mm path length rotating quartz cell
spinning at a rate ensuring that each excitation pulse hits a fresh sample
was used for measurements. The mutual orientation of the excitation
and probe beams was set to the magic angle (54.7ꢀ).
Continuous Irradiation. The sample dissolved in CH2Cl2 was placed
in a sealed 1 cm sample quartz cell. The measurements were performed
in a laser kinetic spectrometer LKS 20 (Applied Photophysics, U.K.)
where the samples were continuously irradiated by a 250 W Xe lamp.
A glass optical filter between the lamp and the irradiated cell was used
to cut off the short wavelength components (λ < 370 nm) to prevent
photodegradation of sample by UV light (this alignment is “light ON”).
The transmitted light passed through a monochromator and was
detected by an R928 photomultiplier (Hamamatsu). Irradiation was
switched off by the addition of a 500 nm long pass filter between the
lamp and the sample cell (this alignment is “light OFF”). In both cases
transmitted light >500 nm served to probe the photochemical release of
gases at the band maxima of the boron cluster adduct. The cycles light
ON/light OFF were periodically repeated.
(6) (a) Olmstead, M. M.; Hope, H.; Benner, L. S.; Balch, A. L. J. Am.
Chem. Soc. 1977, 99, 5502. (b) Bhaduri, S.; Casella, L.; Ugo, R.; Raithby,
P. R.; Zuccaro, C.; Hursthouse, M. B. J. Chem. Soc., Dalton Trans.
1979, 1624.
(7) Tyeklar, Z.; Jacobson, R. R.; Wei, N.; Murthy, N. N.; Zubieta, J.;
Karlin, K. D. J. Am. Chem. Soc. 1993, 115, 2677.
(8) Vaartstra, B. A.; Xiao, J. L.; Cowie, M. J. Am. Chem. Soc. 1990,
112, 9425.
(9) Cheek, Y. M.; Kennedy, J. D.; Thornton-Pett, M. Inorg. Chim.
Acta 1985, 99, L43.
(10) (a) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; ; Cheeseman, J. R.; Jr., J. A. M.; Vreven, T.; Kudin, K. N.;
Burant, J. C.; J. M.Millam; Iyengar, S. S.; Tomasi, J.; Barone, V.;
Mennucci, B.; ; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.;
Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa,
J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.;
Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.;
Jaramillo, J.; ; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.;
Cammi, R.; ; Pomelli, C.; Ochterski, J.; Ayala, P. Y.; Morokuma, K.;
Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich,
S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.;
Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.;
Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.;
Piskorz, P.; ; Komaromi, I.; L.Martin, R.; Fox, D. J.; Keith, T.; Al-Laham,
M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.;
Johnson, B. G.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A.
GAUSSIAN 03, Rev. C.02; Gaussian, Inc: Wallingford, CT, 2004.
(b) Becke, A. D. J. Chem. Phys. 1993, 98, 5648. (c) Lee, C.; Yang, W.;
Parr, R. G. Phys. Rev. B 1988, 37, 785. (d) Hehre, W. J.; Ditchfield, R.;
Pople, J. A. J. Chem. Phys. 1972, 56, 2257. (e) Hay, P. J.; Wadt, W. R.
J. Chem. Phys. 1985, 82, 299.
’ ASSOCIATED CONTENT
S
Supporting Information. Additional information includes
b
details of the DFT calculations mentioned in the text and a listing
of DFT calculated atomic coordinates, a detailed analysis of the 11B
NMR spectra, Figures S1 to S11, and tables of interatomic distances
and angles for compounds 8to 13 obtained from single-crystal X-ray
diffraction data. This material is available free of charge via the
Internet at http://pubs.acs.org.
(11) Le Guennic, B.; Jiao, H.; Kahlal, S.; Saillard, J. Y.; Halet, J. F.;
Ghosh, S.; Shang, M.; Beatty, A. M.; Rheingold, A. L.; Fehlner, T. P.
J. Am. Chem. Soc. 2004, 126, 3203.
(12) Bould, J.; Kennedy, J. D. Chem. Commun. 2008, 2447.
(13) Mikꢁsovskꢀa, J.; Norstrom, J.; Larsen, R. W. Inorg. Chem. 2005,
44, 1006.
’ AUTHOR INFORMATION
Corresponding Author
*E-mails: jbould@gmail.com (J.B.), lang@iic.cas.cz (K.L.).
(14) Zhou, X.; Pan, Q.-H.; Li, M.-X.; Xia, B.-H.; Zhang, H.-X. J. Mol.
Struct. 2007, 822, 65.
’ ACKNOWLEDGMENT
(15) Lang, K.; Mosinger, J.; Wagnerovꢀa, D. M. Coord. Chem. Rev.
2004, 248, 321.
This work was supported by the Czech Science Foundation
(Nos. P207/11/1577, P208/10/1678), the Grant Agency of the
Academy of Sciences of the Czech Republic (KAN400480701),
and the Spanish Ministry of Science Innovation (projects CSD2009-
50, CTQ2009-10132, and a sabbatical leave, SAB2009-0191, to J.B.).
We thank Simon Barrett for assistance with NMR spectroscopy and
Colin Kilner for crystallographic data collection. Aspects of the work
have been supported by personal financial contributions from J.B.
(16) Cavaleiro, J. A. S.; G€orner, H.; Lacerda, P. S. S.; MacDonald,
J. G.; Mark, G.; Neves, M. G. P. M. S.; Nohr, R. S.; Schuchmann, H.-P.;
von Sonntag, C.; Tomꢀe, A. C. J. Photochem. Photobiol. A 2001, 144, 131.
(17) (a) Valentine, D.; Valentine, J. S. Inorg. Chem. 1971, 10, 393.
(b) Vogler, A.; Kunkely, H. Coord. Chem. Rev. 2006, 250, 1622.
(18) Seip, M.; Brauer, H.-D. J. Photochem. Photobiol. A. 1994, 79, 19.
(19) Seip, M.; Brauer, H.-D. J. Photochem. Photobiol. A. 1993, 76, 1.
(20) Vogler, A.; Kunkely, H. J. Am. Chem. Soc. 1981, 103, 6222.
(21) Rupenyan, A.; Commandeur, J.; Groot, M. L. Biochemistry
2009, 48, 6104.
(22) Kapetanaki, S. M.; Field, S. J.; Hughes, R. J. L.; Watmough, N. J.;
Liebl, U.; Vos, M. H. Biochim. Biophys. Acta, Bioenerg. 2008, 1777, 919.
(23) (a) Sortino, S. Chem. Soc. Rev. 2010, 39, 2903. (b) Crespy, D.;
Landfester, K.; Schubert, U. S.; Schiller, A. Chem. Commun. 2010,
46, 6651. (c) Alberto, R.; Motterlini, R. Dalton Trans. 2007, 1651.
(24) Parshall, G. W. Inorg. Synth. 1970, 12, 27.
(25) Wrackmeyer, B.; Kennedy, J. D. J. Magn. Reson. 1980, 38, 529.
(26) (a) Kennedy, J. D. Multinuclear NMR; Mason, J., Ed.; Plenum:
New York and London, 1987; p 221. (b) Reed, D. Chem. Soc. Rev. 1993,
22, 109. (c) Heꢁrmꢀanek, S. Chem. Rev. 1992, 92, 325. (d) Ferguson, G.;
Kennedy, J. D.; Fontaine, X. L. R.; Faridoon; Spalding, T. R. J. Chem. Soc.,
Dalton Trans. 1988, 2555.
(27) McFarlane, W. Proc. R. Soc. (London), Series A 1968, 306, 185.
(28) Merrick, J. P.; Moran, D.; Radom, L. J. Phys. Chem. A 2007,
111, 11683.
ꢁ
The authors would like to thank to Pavel Chꢀabera and Vꢀaclav Slouf
(University of South Bohemia, Novꢀe Hrady) for technical help with
femtosecond experiments.
’ REFERENCES
(1) (a) Bould, J.; McInnes, Y. M.; Carr, M. J.; Kennedy, J. D. Chem.
Commun. 2004, 2380. (b) Bould, J.; Kilner, C. A.; Kennedy, J. D. Dalton
Trans. 2005, 1574.
(2) (a) Maneiro, M.; Fernandez, B.; Gomez-Forneas, E.; Rodriguez,
M. J.; Pedrido, R.; Romero, M. J.; Bermejo, M. R. Z. Anorg. Allg. Chem.
2005, 631, 2000. (b) Vaska, L.; Bath, S. S. J. Am. Chem. Soc. 1966,
88, 1333.
(3) (a) Benner, L. S.; Olmstead, M. M.; Hope, H.; Balch, A. L.
J. Organomet. Chem. 1978, 153, C31. (b) Albrecht, M.; Gossage, R. A.;
Lutz, M.; Spek, A. L.; van Koten, G. Chem.—Eur. J. 2000, 6, 1431.
(4) Pringle, P. G.; Shaw, B. L. J. Chem. Soc., Dalton Trans. 1983, 889.
(5) Balch, A. L.; Benner, L. S.; Olmstead, M. M. Inorg. Chem. 1979,
18, 2996.
(29) (a) COLLECT: Data Collection Strategy Program; Nonius:
Delft, The Netherlands, 1999. (b) Otwinowski, Z.; Minor, W. Methods
7522
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523

Inorganic Chemistry
ARTICLE
Enzymol. 1997, 276, 307. (c) SHELXTL manual; Bruker AXS: Madison,
WI, 1994 and 1998. (d) Sheldrick, G. Acta Crystallogr., Sect. A. 2008,
64, 112.
(30) Barbour, L. J. J. Supramol. Chem. 2001, 1, 189.
(31) Farrugia, L. J. J. Appl. Crystallogr. 1997, 30, 565.
(32) Sluis, P. v. d.; Spek, C. A. Acta Crystallogr., Sect. A. 1990, 46, 194.
(33) Hurley, J. K.; Sinai, N.; Linschitz, H. Photochem. Photobiol.
1983, 38, 9.
7523
dx.doi.org/10.1021/ic200374k |Inorg. Chem. 2011, 50, 7511–7523