Tetranuclear (borole)rhodium clusters with hydrido and methylthiolato ligands - Structures of [{Rh(C4H4BMe)}4(μ 3-H)(μ3-I)3] and of [{Rh(C4H 4BPh)}2(μ-SMe)(μ3-SMe)]2 (2 Rh-Rh)

Article abstract of DOI:10.1002/ejic.200300630

The heterocubanes [Rh(C₄H₄BR)(μ₃-I)] ₄ (1a: R = Ph; 1b: R = Me) can be hydrogenated in THF (suspension, 1 bar H₂, 20 °C) to give the μ₃-hydrido clusters [{Rh(C₄H₄BR)}₄(μ₃-H) (μ₃-I)₃] (2a,b). The structure of 2b consists of a Rh₄(μ₃-H)(μ₃-I)₃ core with idealized C_3v symmetry and four capping η⁵-borole ligands. The Rh-Rh distances within the Rh₃(μ₃-H) cluster unit amount to 2.9448(6), 2.9647(6), and 2.9724(6) A; the μ₃-H ligand is 0.064(5) A above the Rh₃ plane with a mean Rh-H bond length of 1.83(5) A. The tris(acetonitrile) salts [Rh(C₄H₄BR)(NCMe)₃]BF₄ (4a,b) react with NaSMe to give the methylthiolato complexes [Rh(C₄H ₄BR)(SMe)]_x (3a,b). In the crystal 3a possesses a centrosymmetric stepped-ladder structure [{Rh(C₄H₄BPh)} ₂(μ₃-SMe)]₂ (2Rh-Rh) (3a-Rh₄), with a Rh-Rh bond length of 2.9233(11) A. In solution a dissociation equilibrium is observed with a predominating dinuclear complex [Rh(C ₄H₄BPh)(μ-SMe)]₂ (3a-Rh₂); in pyridine a closely related solvate [Rh(py)(C₄H₄BPh)(μ- SMe)]₂ (5) with a static Rh₂(μ-SMe)₂ core is formed. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinstein, Germany, 2004.

Full text of DOI:10.1002/ejic.200300630

FULL PAPER
Tetranuclear (Borole)rhodium Clusters with Hydrido and Methylthiolato
Ligands ؊ Structures of [{Rh(C4H4BMe)}4(µ3-H)(µ3-I)3] and of
[{Rh(C4H4BPh)}2(µ-SMe)(µ3-SMe)]2 (2 Rh؊Rh)[‡]
Gerhard E. Herberich,*[a] Hartmut J. Eckenrath,[a,b] Trixie Wagner,[a,c] and Ruimin Wang[a]
Keywords: Boron / Borole / Hydride ligands / Rhodium / S ligands
The heterocubanes [Rh(C4H4BR)(µ3-I)]4 (1a: R = Ph; 1b: R =
Me) can be hydrogenated in THF (suspension, 1 bar H2, 20
°C) to give the µ3-hydrido clusters [{Rh(C4H4BR)}4(µ3-H)(µ3-
I)3] (2a,b). The structure of 2b consists of a Rh4(µ3-H)(µ3-I)3
core with idealized C3v symmetry and four capping η5-borole
ligands. The Rh−Rh distances within the Rh3(µ3-H) cluster
the methylthiolato complexes [Rh(C4H4BR)(SMe)]x (3a,b). In
the crystal 3a possesses a centrosymmetric stepped-ladder
structure [{Rh(C4H4BPh)}2(µ-SMe)(µ3-SMe)]2 (2Rh−Rh) (3a-
˚
Rh4), with a Rh−Rh bond length of 2.9233(11) A. In solution
a dissociation equilibrium is observed with a predominating
dinuclear complex [Rh(C4H4BPh)(µ-SMe)]2 (3a-Rh2); in pyr-
idine a closely related solvate [Rh(py)(C4H4BPh)(µ-SMe)]2 (5)
˚
unit amount to 2.9448(6), 2.9647(6), and 2.9724(6) A; the µ3-
˚
H ligand is 0.064(5) A above the Rh3 plane with a mean Rh−H with a static Rh2(µ-SMe)2 core is formed.
˚
bond length of 1.83(5) A. The tris(acetonitrile) salts
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
[Rh(C4H4BR)(NCMe)3]BF4 (4a,b) react with NaSMe to give
Germany, 2004)
Introduction
complexes [Rh(SMe)(C4H4BR)]x (3a,b). In the case of 3b a
tetranuclear complex with a partially opened structure is
The recently described heterocubanes [Rh(C4H4BR)(µ3- found in the crystal and a dissociationϪassociation equilib-
I)]4 (1a,b) are readily accessible[2a] and display a rich reac- rium in solution.
tion chemistry.[2b] They possess a total of 72 valence elec-
trons, with five electrons counted per µ3-iodide ligand.
Hence the Rh centers obey the 18-electron rule and there-
fore the complexes 1 are cage compounds. As they have no
Results and Discussion
metalϪmetal bonds, they are not clusters in the sense of the
Hydrogenation and Synthesis of Hydrido Clusters
widely accepted definition of a cluster given by Cotton.[3]
The heterocubanes 1 readily react with dihydrogen in
THF (1 bar, 20 °C, 3 h) to produce the µ3-hydrido clusters
2 (Scheme 1) The basicity of the THF seems to be essential
since no reaction is observed when dichloromethane is used
as solvent. The products can be isolated as dark red crystals
by cooling concentrated dichloromethane solutions. In
solution they are slightly sensitive to air. They do not react
with triethylamine or with iodomethane at ambient tem-
perature and are decomposed by CF3CO2D in CDCl3. We
mention in passing that attempted hydrogenation of 1a,b
with NaBHEt3 was unsuccessful and seemed to result in
reductive decomposition.
The heterocubanes 1a and 1b undergo an exchange of (bor-
ole)rhodium vertices which is fast on the NMR timescale
at ambient temperature,[2a] possibly via a dissociationϪ
association equilibrium between tetra- and dinuclear spec-
ies.
We have tried to vary the bridging ligand in these cage
compounds in a number of ways,[4] and report here on two
results. Homogeneous hydrogenation of the heterocubanes
1a,b (R ϭ Ph and Me, respectively) produces novel tetranu-
clear µ3-hydrido clusters [{Rh(C4H4BR)}4(µ3-H)(µ3-I)3]
(2a,b), and treatment with NaSMe affords the methylthio
[‡]
Borole Derivatives, 25. Part 24: Ref.[1]
[a]
Institut für Anorganische Chemie, Technische Hochschule
Aachen
52056 Aachen, Germany
Fax: (internat.) ϩ 49-241-809-2288
E-mail: gerhard.herberich@ac.rwth-aachen.de
[b]
Convertex Chemie GmbH, Chemie-Park Bitterfeld-Wolfen
06766 Wolfen, Germany
[c]
Novartis Pharma AG, Central Technologies
Scheme 1. Hydrogenation of the heterocubanes [Rh(C4H4BR)(µ3-
I)]4 (1a,b)
4002 Basel, Switzerland
1396
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/ejic.200300630
Eur. J. Inorg. Chem. 2004, 1396Ϫ1401

Tetranuclear (Borole)rhodium Clusters with Hydrido and Methylthiolato Ligands
FULL PAPER
The 1H NMR spectra of the product complexes 2a,b dis-
The RhϪRh distances between the apical Rh atom and
˚
play a quadruplet at high field [e.g. for 2a: δ ϭ Ϫ10.27 ppm, the basal Rh centers amount to more than 3.90 A and thus
1JRh,H ϭ 21.5 Hz] indicating the presence of a triply bridg- are nonbonding. The basal Rh atoms form a nearly regular
˚
ing hydrido ligand in a Rh3(µ3-H) arrangement. The triangle with RhϪRh distances of 2.9448(6)Ϫ2.9724(6) A.
Rh3(µ3-H) structural element was first found in the Fischer/ Known complexes with all-18e configurations and two
Wawersik complex [(RhCp)3(µ3-H)(µ3-Cp)] [δ
ϭ
center, two electron RhϪRh bonds typically display
Ϫ12.47 ppm, 1JRh,H ϭ 26.5 Hz],[5] and is known from a few RhϪRh bond lengths of 2.72 A,[5,6] while electron-deficient
˚
other structurally characterized complexes,[6,7] including, examples[7] tend to have longer bonds, with RhϪRh dis-
[9]
˚
for instance, the tetranuclear complex [(RhCp*)4(µ3- tances up to about 3.0 A.
[7b,7c]
˚
H)4](BF4)2
and the trimethylenemethane complex
The RhϪI bond lengths average 2.737
A
[{Rh(COD)}3(µ3-H){µ3-C(CH2)3}].[7a] We also see two [2.7170(5)Ϫ2.7626(6) A] and are similar to those in 1a and
˚
[3]
˚
types of borole ligands in a 3:1 ratio. In combination with 1b [2.7049(9)Ϫ2.777(1) A], while the RhϪ(µ2-I) distances
the elemental analysis these observations define the formula [2.715Ϫ2.741 A] seem to be marginally shorter.[10] The bor-
˚
and constitution of the clusters 2.
ole ligation to the metal is not detailed here. We only note
that the RhϪB distances [2.307(6)Ϫ2.348(6) A] are in the
˚
expected range (cf. the RhϪB bond lengths in the heterocu-
Structure of the Hydrido Cluster 2b
banes 1a,b, which average to 2.325 A[2a]).
˚
Because of the novelty of these hydrido compounds we
decided to characterize 2b by an X-ray structure determi-
nation. Complex 2b crystallizes in the monoclinic space
group P21/n with Z ϭ 4 (Figure 1; produced with the Platon
software package[8]). The central core of 2b is a trigonal
Rh4(µ3-H)(µ3-I)3 unit with idealized C3v symmetry, and
each Rh vertex is capped by a borole ligand. Note that the
apical vertex (with Rh1) is similar to the vertices of 1b,
while the basal vertices belong to a true cluster unit (µ3-
H)Rh3.
The hydride ligand could be localized in the structure
analysis and was refined isotropically. It lies 0.64(5) A above
˚
the Rh3 plane, and the RhϪH bond lengths observed aver-
˚
age to 1.83(5) A. This result is in reasonable agreement with
the data for the cluster [(RhCp*)4(µ3-H)4](BF4)2 (as deter-
mined by neutron diffraction).[7b]
The hydrido clusters [{Rh(C4H4BR)}4(µ3-H)(µ3-I)3]
(2a,b) possess a total of 68 valence electrons, four less than
the cage complexes 1. Formally removing one iodo ligand
from the cages 1 generates a [{Rh(C4H4BR)}4(µ3-I)3]ϩ frag-
ment with three empty, metal-centered orbitals ready for σ-
bonding. In C3v symmetry these σ-orbitals combine to an
a1 and an e set of orbitals. Interaction with the 1s orbital
of the hydrido ligand results in bonding of the hydrido li-
gand and metal-metal bonding, in other words, in a four
center, two electron bond within the (µ3-H)Rh3 tetra-
hedron. The e set of MOs remains empty and accounts for
the reduced electron count of 2a,b.
Synthesis of Methylthiolato Complexes
The anionic ligand of the cage complexes 1 may be sub-
stituted in a two-step procedure, first removing the halide
ligand by dehalogenation with silver salts such as AgBF4 in
acetonitrile, and then introducing a new ligand. Thus, the
reaction of the tris(acetonitrile) salts 4a,b with sodium
methylthiolate in acetonitrile smoothly affords the robust,
high-melting methylthiolato complexes 3a,b (Scheme 2).
The phenyl complex 3a can be crystallized from hot tolu-
ene, while the methyl complex 3b is scarcely soluble, even
in toluene or CH2Cl2.
Figure 1. Molecular structure (PLATON plot,[8] at the 30% prob-
ability level) of 2b in the crystal; see text for RhϪRh interactions in
˚
the cluster; selected bond lengths (A) and bond angles (°): I1ϪRh1 Structure of the Methylthiolato Complex 3a in the Crystal
2.7420(5), I1ϪRh2 2.7181(6), I1ϪRh3 2.7309(6), I2ϪRh1
The low-temperature NMR spectra of complex 3a indi-
cate that this compound is not of the heterocubane type.
2.7272(6), I2ϪRh3 2.7170(5), I2ϪRh4 2.7285(6), I3ϪRh1
2.7626(6), I3ϪRh2 2.7398(6), I3ϪRh4 2.7626(6), Rh1···Rh2
3.956(1), Rh1··Rh3 3.905(1), Rh1···Rh4 3.973(1), Rh2ϪRh3
2.9647(6), Rh2ϪRh4 2.9448(6), Rh3ϪRh4 2.9724(6), Rh2ϪH1
1.82(5), Rh3ϪH1 1.82(5), Rh4ϪH1 1.83(5); I1ϪRh1ϪI2 87.81(2),
I1ϪRh1ϪI3 85.87(2), I1ϪRh2ϪI3 86.78(2), I1ϪRh3ϪI2 88.24(2),
I2ϪRh1ϪI3 86.17(1), I2ϪRh4ϪI3 86.14(1), Rh1ϪI1ϪRh2
92.85(2), Rh1ϪI1ϪRh3 91.06(2), Rh1ϪI2ϪRh3 91.68(2),
Rh1ϪI2ϪRh4 93.47(2), Rh1ϪI3ϪRh2 91.93(2), Rh1ϪI3ϪRh4
91.95(1), Rh2ϪI1ϪRh3 65.92(1), Rh2ϪI3ϪRh4 64.71(2), Scheme 2. Synthesis of the methylthiolato complexes
Rh3ϪI2ϪRh4 66.16(1)
[Rh(SMe)(C4H4BR)]x (3a,b)
Eur. J. Inorg. Chem. 2004, 1396Ϫ1401
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1397

G. E. Herberich, H. J. Eckenrath, T. Wagner, R. Wang
FULL PAPER
The X-ray single-crystal structure determination revealed structures. For instance, while the chloro complex
that complex 3a consists of tetranuclear [{Rh- [CuCl(PPh3)]4 is a heterocubane, the phenylthiolate adopts
(C4H4BR)}2(µ-SMe)(µ3-SMe)]2 (2 RhϪRh) (3a-Rh4) a [{Cu(PPh3)}2(µ-SPh)(µ3-SPh)]2 structure closely akin to
molecules which display a crystallographic center of sym- that of 3a. This structural difference is probably related to
metry (Figure 2). The structure observed may formally be the bulkiness of the phenylthiolato ligands.
˚
deduced from a cubanoid structure analogous to 1a. Open-
ing the heterocubane cage gives a stepped ladder, and coor-
dinative unsaturation of the outer Rh centers is avoided by
formation of dative RhϪRh bonds [Rh1ϪRh2 2.9233(11)
The RhϪS bond lengths in 3a (2.32Ϫ2.46 A) are shorter
˚
than the RhϪI bond lengths in 1a (2.73Ϫ2.77 A). The
opening of the cubanoid structure to the stepped-ladder
structure may again be caused by steric congestion. A de-
tailed molecular modeling study would require a fully para-
metrized force-field to deal with the η5-bonded phenylbor-
ole ligands. We therefore recurred to the following simpli-
fied approach for the generation of a hypothetical µ3-SMe
bridged heterocubane: Starting with the structure of 1a the
distances between the Rh centers and the heteroatoms were
˚
A]. Some additional stabilization by π-interactions with the
sulfur lone-pairs may also be present. While the central
rhombus is planar (as implied by the centrosymmetry of
the molecule), the outer tetragons are folded along the line
Rh1,Rh2 with
a folding angle (between the plane
Rh1,Rh2,S1 and Rh1,Rh2,S2) of 68.38(9)° in such a way
that S2 is moved toward the central rhombus. Finally, we
note that the molecule 3a-Rh4 may be viewed as a dimer of
a dinuclear species 3a-Rh2. This view is supported by the
fact that the connecting Rh2ϪS1Ј and Rh2ЈϪS1 bonds are
reduced to 2.35 A (for pertinent reference data see ref.[10]),
and the methyl groups were placed at an SϪC distance of
1.8 A in the radial direction. No conformational changes
and no optimizations of the whole structure were per-
formed. The resulting molecule exhibited prohibitively
short C···H (1.9 A), C···C (2.8 A), and H···H (1.9 A) inter-
actions between the SϪMe and BϪPh groups. On the basis
of MM2 van der Waals parameters[12] an estimated destabi-
lization of the cubanoid structure by approximately 100 kJ/
mol was calculated.
˚
˚
˚
markedly longer [2.461(2) A] than the other RhϪS bonds
˚
˚
˚
in the same molecule [cf. Rh1ϪS1 2.326(2), Rh1ϪS2
˚
˚
2.303(3) A and Rh2ϪS1 2.400(2), Rh2ϪS2 2.405(3) A].
Structure of the Methylthiolato Clusters 3a,b in Solution
The low-temperature 1H NMR spectrum of 3a
(500 MHz, CD2Cl2, Ϫ80 °C) displays eight multiplets of
equal intensity for the borole protons. This observation
strongly suggests that the same structure that was found in
the crystal is also present in the low-temperature solution.
The stepped-ladder molecule 3a-Rh4 is a meso species with
two chemically different borole ligands, both of which are
in a chiral environment. Two sharp singlets are seen for the
SMe groups, again in agreement with the assumption of a
stepped-ladder structure in solution.
1H NMR measurements at ambient temperature or above
(in CD2Cl2, CDCl3, or 1,1,2,2-C2D2Cl4) produced spectra
corresponding to effective C2v symmetry. Using a 500 MHz
Figure 2. Molecular structure (PLATON plot,[8] at the 30% prob- spectrometer the signals were broad and blurred; on an
˚
ability level) of 3a in the crystal; selected bond lengths (A) and
80 MHz machine the same solutions gave sharp, well-re-
bond angles (°):Rh1ϪRh2 2.9233(11), Rh1ϪS1 2.326(2), Rh1ϪS2
solved signals with the characteristic hyperfine structure of
the borole AAЈBBЈ four-spin system. Only one type of bor-
ole ligand and one type of SMe ligand were left. The chemi-
cal shifts are temperature dependent up to about 80 °C and
then remain constant up to 120 °C (in 1,1,2,2-C2D2Cl4), the
highest temperature reached. These observations are readily
explained by the assumption of a dissociation/association
equilibrium between the tetranuclear species 3a-Rh4 and
the dinuclear species 3a-Rh2, which was already mentioned
above in the structural discussion (Scheme 3). We mention
in passing that dinuclear species such as 3a-Rh2 could, in
principle, form three stereoisomers, two syn-stereoisomers
2.303(2), Rh2ϪS1 2.400(2), Rh2ϪS2 2.405(3), Rh2ϪS1Ј 2.461(2),
S1ϪC1 1.818(8), S2ϪC2 1.826(10); Rh1ϪRh2ϪS1 50.67(5),
Rh1ϪRh2ϪS2 50.06(6), Rh2ϪRh1ϪS1 52.92(5), Rh2ϪRh1ϪS2
53.21(7), S1ϪRh1ϪS2 82.78(8), S1ϪRh2ϪS2 79.14(8)
Opening of heterocubane structures is also known from and one anti-stereoisomer, although only one species is seen
the family of complexes [MX(PR3)]4,[11] and small metal in the NMR spectra. A comparison of the chemical shifts
centers and bulky counter-ligands favor stepped-ladder of the tetranuclear species from the spectrum measured at
1398
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 1396Ϫ1401

Tetranuclear (Borole)rhodium Clusters with Hydrido and Methylthiolato Ligands
FULL PAPER
Ϫ80 °C and of the dinuclear species 3a-Rh2 exclusively pre- ible borole complexes,[14] and therefore have been the start-
sent above 80 °C shows that at ambient temperatures 3a- ing point of a variety of studies.[1,2,15] In this paper we have
Rh2 is already the predominant species. It is interesting to described the first true cluster compounds with (borole)rho-
note here that the electronically related ruthenium com- dium vertices. In preceding work we speculated that the fast
plexes [(RuCp*)2(µ-SR)2] are dinuclear, as confirmed by an exchange of (borole)rhodium vertices observed for the het-
X-ray structure determination of [{Ru(C5Me4Et)}2(µ- erocubanes 1a,b could occur via a dissociative formation
SEt)2].[13]
of dinuclear intermediates.[2a] It was therefore particularly
satisfying to find compelling evidence for the existence of
the dinuclear species 3a,b-Rh2 in equilibrium with the tetra-
nuclear clusters 3a,b-Rh4.
Experimental Section
Sdheme 3. The dissociation equilibrium of the tetranuclear com-
General: Reactions were carried out under an atmosphere of dini-
trogen by means of conventional Schlenk techniques. Hexane was
distilled from potassium, CH2Cl2 from CaH2, and THF and Et2O
from sodium benzophenone ketyl.
plex [{Rh(C4H4BR)}2(µ-SMe)(µ3-SMe)]2 (2 RhϪRh) (3a-Rh4)
In the presence of pyridine complex 3a shows a some-
what enhanced solubility, and in neat [D5]pyridine it is
moderately soluble. The 1H NMR spectrum of this solution
shows effective C2v symmetry. The SMe groups appear as a
sharp triplet not seen in the equilibrium mixture of
Scheme 3. The triplet results from coupling with two Rh
centers and indicates the formation of a new dinuclear spec-
Secondary ion mass spectra (SIMS) were recorded on a Finnigan
MAT-95 spectrometer. NMR spectra were recorded on a Varian
Unity 500 spectrometer (1H: 499.6 MHz; 13C{1H}: 125.6 MHz;
11B{1H}: 160.3 MHz). Chemical shifts are given in ppm; they were
measured at ambient temperature and are relative to internal TMS
1
for H and 13C and relative to BF3·OEt2 as external reference for
ies 5, presumably of the formula [{Rh(py)(C4H4BPh)}2(µ- 11B.
SMe)2]. On attempted isolation complex 5 lost the pyridine
[{Rh(C4H4BPh)}4(µ3-H)(µ3-I)3]
(2a):
The
heterocubane
completely, re-forming the cluster 3a. The observation of
the sharp triplet demonstrates that 3a-Rh2 cannot dis-
sociate into a mononuclear species, not even in the presence
of a good donor ligand. On the other hand, it also suggests
that the tetranuclear species 3a-Rh4 is no longer present as
it would, very likely, cause blurring of the hyperfine struc-
ture of the signal.
[Rh(C4H4BPh)(µ3-I)]4 (1a; 830 mg, 0.561 mmol) was suspended in
THF (60 mL). A slow stream of hydrogen was passed through the
flask with stirring for 3 h. Complex 1b slowly dissolved as the reac-
tion proceeded. The volatiles were then removed under vacuum.
The residue was dissolved in CH2Cl2 (50 mL); the solution was
filtered through a frit covered with kieselguhr, concentrated to a
small volume (5 mL), and kept at Ϫ30 °C for 24 h. After removal
of the mother liquor the crystals were washed with diethyl ether (2
ϫ 5 mL) and dried under vacuum to give 2a (560 mg, 74%) as dark
red crystals, m.p. 212 °C (dec.); in solution somewhat sensitive to
air, soluble in CH2Cl2, slightly soluble in benzene, insoluble in hex-
ane, Et2O and acetonitrile. C40H37B4I3Rh4 (1353.3): calcd. C 35.50,
H 2.76; found C 35.18, H 2.71. SIMS (NBA; negative ions): m/z
(%) ϭ 1110 (14) [Rh3I3(C4H4BPh)3]Ϫ, 497 (24) [RhI2(C4H4BPh)]Ϫ,
357 (27) [RhI2]Ϫ, 127 (100) [IϪ]. 1H NMR (CDCl3): δ ϭ Ϫ10.27
1
3
4
(q, JRh,H ϭ 21.5 Hz, µ3-H), 3.84 (m, N ϭ J23 ϩ J24 ϭ 5.5 Hz,
borole 2-/5-H), 4.49 (m, N ϭ 5.5 Hz, 3 ϫ borole 2Ј-/5Ј-H), 4.91 (m,
N ϭ 5.2 Hz, borole 3-/4-H), 5.11 (m, N ϭ 5.2 Hz, 3 ϫ borole 3Ј-/
4Ј-H), 7.31Ϫ7.39 (m, Hp ϩ 3 HpЈ ϩ 2 Hm ϩ 6 HmЈ), 7.57 (m, 2
Ho), 7.62 (m, 6 HoЈ) ppm. 13C{1H} NMR (CDCl3): δ ϭ 71.8 (br.,
The analogous 1-methylborole complex 3b poses the
1
same structural problems. Because of the low solubility of
borole C-2,5), 75.0 (br., borole C-2Ј,5Ј), 88.49 [d, JRh,C ϭ 9.3 Hz,
this complex only 1H NMR spectra could be measured and borole C-3,4], 90.19 (d, JRh,C ϭ 8.3 Hz, borole C-3Ј,4Ј) 127.68
1
(Cm), 127.60 (CmЈ), 129.09 (CpЈ), 129.11 (Cp), 135.82 (Co ϩ CoЈ),
low-temperature measurements were not possible. Com-
parison of 1H NMR measurements at 500 MHz and at
80 MHz revealed chemical-shift differences, suggesting that
the methyl compound 3b is also involved in a dissociation
equilibrium in solution. An attempted X-ray structure de-
termination showed the crystals to be polysynthetic twins,
and no structure solution could be obtained. Thus the nu-
clearity of 3b remains an open question.
136.7 (Ci), 137.08 (CiЈ) ppm. 11B{1H} NMR (CDCl3): δ ϭ 20 ppm.
[{Rh(C4H4BMe)}4(µ3-H)(µ3-I)3]
(2b):
The
heterocubane
[Rh(C4H4BMe)(µ3-I)]4 (1b, 200 mg, 0.162 mmol) was dissolved in
THF (20 mL) and hydrogenated as described above for 1a. The raw
product was dissolved in CH2Cl2 (10 mL); the solution was filtered
through a frit covered with kieselguhr, concentrated to a small vol-
ume (1 mL), and kept at Ϫ30 °C for 24 h. After removal of the
mother liquor the crystals were washed with hexane (2 ϫ 5 mL)
and dried under vacuum to give 2b (148 mg, 82%) as dark red crys-
tals, m.p. 155 °C (dec.); in solution somewhat sensitive to air, sol-
uble in CH2Cl2, tetrahydrofuran and toluene, moderately soluble
Conclusion
The triple-decker complexes [{Rh(C4H4BR)}2(µ-
C4H4BR)] (R ϭ Ph, Me) are among the most easily access- in Et2O, insoluble in hexane. Crystals suitable for a single-crystal
Eur. J. Inorg. Chem. 2004, 1396Ϫ1401
www.eurjic.org  2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim 1399

G. E. Herberich, H. J. Eckenrath, T. Wagner, R. Wang
FULL PAPER
Table 1. Crystal data, data collection parameters, and convergence results for 2b and 3a
2b
3a
Empirical formula
Molecular mass
Crystal system
Space group
C20H29B4I3Rh4
1105.01
monoclinic
P21/n
Mo-Kα (0.71073)
15.415(2)
11.629(2)
15.834(2)
104.769(10)
2744.6(7)
4
C44H48B4Rh4S4
1159.94
orthorhombic
Pbca
Cu-Kα (1.54184)
8.408(4)
22.058(7)
˚
Radiation (λ[A])
˚
a (A)
˚
b (A)
˚
c (A)
23.915(7)
β (°)
3
˚
V (A )
Z
4435(3)
4
1.737
2304
13.819
dcalcd. [g/cm3]
F(000)
2.67
2032
5.746
µ [cmϪ1
]
Absorption correction
Max./min. transmission
Θ (°)
empirical
0.9986/0.8981
3Ϫ28
empirical
0.9976/0.7565
5Ϫ65
Temperature [K]
203
293
Scan mode
ωϪ2Θ
ωϪΘ
0.35 ϫ 0.15 ϫ 0.06
3707
3707
2460
I Ͼ 2σ(I)
255
0.0491 (0.0958)
0.1228(0.1355)
1.008
Crystal size [mm]
Reflections collected
Reflections unique
Reflections observed
Criterion for observation
Variables
0.36 ϫ 0.28 ϫ 0.20
8687
6609[a]
5233
I Ͼ 2σ(I)
288
0.0286 (0.0504)
0.0567 (0.0598)
1.010
R1,[b] observed (all data)
wR2,[c] observed (all data)
GOF[d]
3
˚
Max. residual density [e/A ]
0.89
1.495
[a]
[b]
[c]
2
2
2
2
Rint ϭ 0.0236. R1 ϭ Σ||Fo| Ϫ |Fc||/Σ|Fo|. wR2 ϭ [Σw(Fo Ϫ Fc )2/Σw(Fo2)2]1/2, where w ϭ 1/[σ2(Fo ) ϩ (aP)2] and P ϭ [max(Fo ,0)
2
[d]
2
2 2
ϩ 2Fc ]/3. GOF ϭ [Σw(Fo Ϫ Fc ) /Σ[(n Ϫ p)]1/2
.
structure determination were obtained by very slowly concentrating
a saturated solution of 2b in diethyl ether. C20H29B4I3Rh4 (1105.5): 4.17; found C 45.93, H 4.16. H NMR (80 MHz, CDCl3, 25 °C):
to ambient temperature. C44H48B4Rh4S4 (1160.0) calcd. C 45.56, H
1
calcd. C 21.74, H, 2.65; found C 21.69, H 2.67. SIMS (NBA; posi-
δ ϭ 2.70 (s, Me), 3.96 (m, N ϭ 5.8 Hz, borole 2-/5-H), 4.60 (m,
tive ions): m/z (%) ϭ 1106 (62) [Mϩ], 979 (18) [Mϩ Ϫ I], 798 (100) N ϭ 4.6 Hz, borole 3-/4-H), 7.24Ϫ7.43 (complex region, 2 Hp ϩ 4
[Mϩ Ϫ I Ϫ Rh Ϫ C4H4BMe]; negative ions: m/z (%) ϭ 435 (26) Hm), 7.73 (m, 4 Ho) ppm. 13C{1H} NMR (125.7 MHz, CDCl3, 25
1
[RhI2(C4H4BMe)]Ϫ, 127 (100) [IϪ]. H NMR (CDCl3): δ ϭ Ϫ9.78
°C): δ ϭ 23.84 (Me), 93.04 (br., borole C-3/4), 127.90 (Cm), 129.34
1
[q, JRh,H ϭ 21.4 Hz, µ3-H), 0.40 (s, Me), 0.50 (s, 3 ϫ MeЈ), 3.41 (Cp), 135.56 (Co) ppm; signal for C-2/5 not observed. 11B{1H}
(m, N ϭ 5.2 Hz, borole 2-/5-H), 4.15 (m, N ϭ 5.2 Hz, 3 ϫ borole
2Ј-/5Ј-H), 4.89 (m, N ϭ 5.5 Hz, borole 3-/4-H), 5.42 (m, N ϭ
5.5 Hz, 3 ϫ borole 3Ј-/4Ј-H) ppm. 13C{1H} NMR (CDCl3): δ ϭ
Ϫ2.31 (br., Me ϩ MeЈ), 72.81 (br., borole C-2,5), 76.15 (br., borole
C-2Ј,5Ј), 87.73 [d, 1JRh,C ϭ 9.3 Hz, borole C-3,4], 89.99 [d, 1JRh,C ϭ
8.7 Hz, borole C-3Ј,4Ј] ppm. 11B{1H} NMR (CDCl3): δ ϭ 25 ppm.
NMR (160.4 MHz, CDCl3, 25 °C): δ ϭ 22 ppm.
Selected VT NMR Data for 3a: 1H NMR (500 MHz, CD2Cl2, Ϫ80
°C): δ ϭ 2.14 (s, µ3-SMe), 2.47 (s, µ-SMe), 2.80 (m, 2 H, borole 2-
H2 or 5-H), 3.23 (m, 2 H, borole 2-H or 5-H), 3.53 (m, borole 2Ј-
H or 5Ј-H), 3.75 (m, borole 2Ј-H or 5Ј-H), 3.92 (m, borole 3Ј-H or
4Ј-H), 4.35 (m, borole 3-H or 4-H), 4.52 (m, borole 3-H or 4-H),
5.46 (m, borole 3Ј-H or 4Ј-H), 7.36Ϫ7.40 (complex region, 2 Hp ϩ
4 Hm), 7.47 (m, 4 HmЈ), 7.53 (m, 2 HpЈ), 7.65Ϫ7.70 (complex region,
4 Ho ϩ 2 HoЈ) ppm; assignments were made on the basis of
[{Rh(C4H4BPh)}2(µ-SMe)(µ3-SMe)]2 (3a): Polymeric catena-
[Rh(µ,η5:η6-C4H4BPh)BF4][1] (650 mg, 1.971 mmol Rh) was dis-
solved in acetonitrile (10 mL) to give a solution of 4a. This solution
was added with stirring to a suspension of NaSMe (138 mg,
1.969 mmol) in acetonitrile (5 mL), and stirring was continued for
60 min. The solid raw product thus obtained was collected on a
G4 frit covered with kieselguhr and washed with acetonitrile (5 ϫ
10 mL). It was then eluted with CH2Cl2. The solvent was evapo-
rated under vacuum, and the resulting solid was washed with ace-
tone (2 ϫ 5 mL) and dried under vacuum to afford 3a (430 mg,
75%) as blackish red microcrystals, m.p. 238 °C, not air-sensitive,
1
(1H,1H)-COSY spectra. H NMR (500 MHz, CD2Cl2, 25 °C): δ ϭ
2.98 (br. s, 2 Me), 4.19 (br. s, 2 ϫ borole 2-/5-H), 4.73 (br. s, 2 ϫ
borole 3-/4-H), 7.33 (m, 4 Hm), 7.42 (m, 2 Hp), 7.74 (m, 4 Ho) ppm.
1H NMR (500 MHz, 1,1,2,2-C2D2Cl4, 120 °C): δ ϭ 3.13 (s, 2 Me),
4.28 (m, 2 ϫ borole 2-/5-H), 4.82 (m, 2 ϫ borole 3-/4-H), 7.28 (m,
4 Hm), 7.38 (m, 2 Hp), 7.71 (m, 4 Ho) ppm.
1
soluble in pyridine, slightly soluble in CHCl3, CH2Cl2 and toluene, Data for 5: H NMR (80 MHz, [D5]pyridine, 25 °C): δ ϭ 2.76 (t,
insoluble in diethyl ether, acetone, acetonitrile. Single crystals suit-
able for a single-crystal structure determination were obtained from H), 5.26 (m, N ϭ 5.0 Hz, borole 3-/4-H), 7.19Ϫ7.32 (complex re-
a concentrated solution of 3a in hot toluene by very slow cooling gion, 4 Hm ϩ 2 Hp), 7.75 (m, 4 Ho) ppm.
3JRh,H ϭ 1.4 Hz, 2 Me), borole 4.09 (m, N ϭ 5.7 Hz, borole 2-/5-
1400
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 1396Ϫ1401

Tetranuclear (Borole)rhodium Clusters with Hydrido and Methylthiolato Ligands
FULL PAPER
[1]
[Rh(SMe)(C4H4BMe)]x (3b):
A solution of AgBF4 (221 mg,
G. E. Herberich, H. J. Eckenrath, U. Englert, Organometallics
1998, 17, 519Ϫ523.
1.135 mmol) in acetonitrile (10 mL) was added with stirring to a
suspension of 1b (350 mg, 0.284 mmol) in acetonitrile (10 mL), and
stirring was continued for 30 min. The resulting reaction mixture
was filtered through a frit covered with kieselguhr into a flask con-
taining a stirred suspension of NaSMe (79.5 mg, 1.134 mmol) in
acetonitrile (5 mL). After stirring the reaction mixture for 60 min,
the microcrystalline product was collected on a frit covered with
kieselguhr and washed with acetonitrile (5 ϫ 10 mL). It was then
eluted with CH2Cl2. The solvent was evaporated under vacuum,
and the resulting solid was washed with acetone (2 ϫ 5 mL) and
dried under vacuum to afford 3b (180 mg, 70%) as blackish red
microcrystals, m.p. 227 °C; not air-sensitive, slightly soluble in
CHCl3, CH2Cl2 and toluene, insoluble in diethyl ether, acetone,
acetonitrile. 1H NMR (80 MHz, CDCl3, 25 °C): δ ϭ 0.31 (s, B-
Me), 2.83 (s, S-Me), 3.77 (m, N ϭ 5.7 Hz, borole 2-/5-H), 5.24 (m,
[2]
[2a] G. E. Herberich, H. J. Eckenrath, U. Englert, Organometall-
ics 1997, 16, 4292Ϫ4298. [2b] G. E. Herberich, H. J. Eckenrath,
U. Englert, Organometallics 1997, 16, 4800Ϫ4806.
[3]
[4]
F. A. Cotton, Quart. Rev. Chem. Soc. 1966, 20, 389Ϫ401.
H. J. Eckenrath, Doctoral Dissertation, Rheinisch-Westfälische
Technische Hochschule Aachen, Aachen, Germany, 1997.
[5a] E. O. Fischer, H. Wawersik, Chem. Ber. 1968, 101, 150Ϫ155.
[5]
[5b]
O. S. Mills, E. F. Paulus, J. Organomet. Chem. 1968, 11,
587Ϫ594.
[6] [6a]
´
T. Teppana, S. Jäärskeläinen, M. Ahlgren, J. Pursiainen, T.
[6b]
A. Pakkanen, J. Organomet. Chem. 1995, 486, 217Ϫ228.
Rossi, J. Pursiainen, M. Ahlgren, T. A. Pakkanen, J. Or-
S.
´
[6c]
ganomet. Chem. 1990, 391, 403Ϫ414.
J. Pursiainen, T. A.
Pakkanen, S. Jäärskeläinen, J. Organomet. Chem. 1985, 290,
85Ϫ98.
[7] [7a]
G. E. Herberich, U. Englert, L. Wesemann, P. Hofmann,
1
N ϭ 5.8 Hz, borole 3-/4-H) ppm. H NMR (500 MHz, CDCl3, 25
°C): δ ϭ 0.32 (s, Me), 2.71 (s, SMe), 3.69 (m, borole 2-/5-H), 5.22
(m, borole 3-/4-H) ppm.
Angew. Chem. 1991, 103, 329Ϫ331; Angew. Chem. Int. Ed.
[7b]
Engl. 1991, 30, 313Ϫ315.
J. S. Ricci, T. F. Koetzle, R. J.
Goodfellow, P. Espinet, P. M. Maitlis, Inorg. Chem. 1984, 23,
[7c]
1828Ϫ1831.
P. Espinet, P. M. Bailey, P. Piraino, P. M.
X-ray Crystal Structure Determinations: The data collection was
performed on an ENRAF-Nonius CAD4 diffractometer with Mo-
Kα radiation for 2b and Cu-Kα radiation for 3a (where the crystals
were small and the unit cell comparatively large) using a graphite
monochromator. Crystal data, data collection parameters, and con-
vergence results are given in Table 1. In the case of 3a all residual
[7d]
Maitlis, Inorg. Chem. 1979, 18, 2706Ϫ2710.
R. R. Burch,
E. L. Muetterties, M. R. Thompson, V. W. Day, Organometall-
ics 1982, 2, 474Ϫ478.
[8]
[9]
PLATON: A. L. Spek, Acta Crystallogr., Sect. A 1990, 46, C34.
Cf. data for [(RhCp*)4(µ3-H)4](BF4)2: two short bonds
[7b,7c]
˚
˚
of 2.655(1) A and four long bonds of 2.829(1) A;
and
3
for the 44e cluster [{Rh(COD)}3(µ3-H){µ3-C(CH2)3}]:
˚
electron density greater than 0.55 e/A was close to the two Rh
atoms.
[7a]
˚
2.977(1)Ϫ2.981(1) A].
[10]
[11]
A. G. Orpen, L. Brammer, F. H. Allen, O. Kennard, D. G.
Watson, R. Taylor, J. Chem. Soc., Dalton Trans. 1989, S1ϪS83.
I. G. Dance, M. L. Scudder, L. J. Fitzpatrick, Inorg. Chem.
1985, 24, 2547Ϫ2550.
N. L. Allinger, J. Am. Chem. Soc. 1977, 99, 8127Ϫ8134.
U. Koelle, C. Rietmann, U. Englert, J. Organomet. Chem. 1992,
423, C20ϪC23.
CCDC-216303 (for 2b) and -216302 (for 3a) contain the sup-
plementary crystallographic data for this paper. These data can
be obtained free of charge at www.ccdc.cam.ac.uk/conts/
retrieving.html [or from the Cambridge Crystallographic Data
Centre, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: (internat.)
ϩ 44-1223-336-033; E-mail: deposit@ccdc.cam.ac.uk].
[12]
[13]
[14] [14a]
G. E. Herberich, B. Heßner, W. Boveleth, H. Lüthe, R.
Saive, L. Zelenka, Angew. Chem. 1983, 95, 1024Ϫ1025; Angew.
Chem. Int. Ed. Engl. 1983, 22, 966; Angew. Chem. Suppl. 1983,
[14b]
1503Ϫ1510.
G. E. Herberich, U. Büschges, B. Heßner, H.
Acknowledgments
We thank Prof. Dr. U. Englert for helpful discussions. This work
was generously supported by the Deutsche Forschungsgemeinsch-
aft and the Fonds der Chemischen Industrie.
Lüthe, J. Organomet. Chem. 1986, 312, 13Ϫ25.
G. E. Herberich, U. Büschges, Chem. Ber. 1989, 122, 615Ϫ619.
Received September 3, 2003
[15]
Early View Article
Published Online February 26, 2004
Eur. J. Inorg. Chem. 2004, 1396Ϫ1401
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1401