Remarkable stability of metallocenes with superbulky ligands: Spontaneous reduction of SmIII to SmII

Article abstract of DOI:10.1002/anie.200705001

Unexpected attraction between extremely bulky cyclopentadienyl ligands of opposite chirality has been observed in perarylated metallocenes [(Ar ₅C₅)₂M]. The exceptional stability of such sterically congested metallocenes (see structure) is explained by a merry-go-round C-H...C(π) hydrogen-bond network. This stabilizing force even enables spontaneous reduction of a Sm^III precursor to a Sm ^II metallocene. (Figure Presented)

Full text of DOI:10.1002/anie.200705001

Angewandte
Chemie
DOI: 10.1002/anie.200705001
Cyclopentadienyl Complexes
Remarkable Stability of Metallocenes with Superbulky Ligands:
Spontaneous Reduction of SmIII to SmII**
Christian Ruspic, John R. Moss, Markus Schürmann, and Sjoerd Harder*
Complexes containing extremely bulky ligands such as the
perphenylated cyclopentadienyl ligand (Ph5C5) have shown
unique properties or coordination geometries.[1] For example,
the typical bent structures of the carbene analogue decame-
thylstannocene ([Cp*2SnII], Cp* = C5Me5) can be linearized
by use of Ph5C5 ligands.[2] Surprisingly, the nickelocene
complex [(Ph5C5)2Ni], prepared by reaction of the Ph5C5C
radical with a Ni0 derivative, is diamagnetic.[3] The existence
of [(Ph5C5)2Fe]was long disputed on the basis of its short
ring–ring distance.[4] Today, two linkage isomers are known:
the more sterically relaxed zwitterionic h5,h6-complex 1[5] and
the sterically congested (h5,h5)-metallocene 2.[6]
of half-sandwich complexes, generally sterically congested
metallocenes are formed.
Complexes containing the Ph5C5 ligand are very insoluble
on account of their rigidity and high symmetry.[1] Therefore,
we used the modified perarylated cyclopentadiene (4-nBu-
C6H4)5C5H (CpBIGH), which can be obtained in a simple high-
yield, one-pot procedure.[10] This ligand can be smoothly
deprotonated by [(4-tBu-benzyl)2(thf)4Ca]in benzene at 60 8C
(Scheme 1). However, instead of heteroleptic [(CpBIG)(4-tBu-
benzyl)Ca], the calcocene [(CpBIG)2Ca]was isolated as large
yellow crystals that are readily soluble in hexane (43% yield
based on CpBIGH). Despite the bulkiness of the CpBIG ligand,
the Schlenk equilibrium is apparently at the homoleptic side.
Reactions of [(CpBIG)2Ca]with [(4- tBu-benzyl)2(thf)4Ca]or
[(2-Me2N-a-Me3Si-benzyl)2(thf)2Ca]both gave no ligand
exchange, which implies high stability of the calcocene.[11]
Reaction of equimolar amounts of CpBIGH with [(2-Me2N-
benzyl)3Y]in benzene (Scheme 2) gave the half-sandwich
complex [(CpBIG)(2-Me2N-benzyl)2Y]as colorless needlelike
crystals (yield: 31%). However, the analogous reaction with
[(2-Me2N-benzyl)3Yb]resulted in spontaneous reduction to
the YbII metallocene [(CpBIG)2Yb], which crystallized as dark
green blocks (yield: 22%; the yield could be increased to 36%
by using the correct 2:1 stoichiometry).
First reports on the use of the Ph5C5 ligand in catalysis
show that it greatly enhances stereoselectivity in asymmetric
syntheses.[7] Moreover, the bulk of the ligand increases the
activity and selectivity in chromium-catalyzed ethylene oli-
gomerization.[8] As half-sandwich complexes of alkaline-earth
and lanthanide metals are effective and promising catalysts
for a wide variety of applications,[9] we pursued the introduc-
tion of superbulky cyclopentadienyl ligands in these com-
pound classes. Our surprising initial results show that instead
Spontaneous reduction of LnIII ions (Ln = lanthanide) is
easily accomplished for Eu (Eu3+/Eu2+
E
1/2 = ꢀ0.35 V)[12] and
has been observed on several occasions.[13] Reduction of YbIII
is less favorable (Yb3+/Yb2+
observed before;
E
1/2 = ꢀ1.15 V)[12] but has been
decomposition
of
the
dimer
II
[{(MeH4C5)2YbMe}2]to the Yb metallocene in toluene is
an extraordinarily slow reaction [Eq. (1)].[14] Reaction times
can be reduced substantially by addition of diethyl ether as a
coordinating solvent (50% conversion is observed after 8 h at
808C in a pressure vessel).[14] The importance of Lewis bases is
underscored by a more recent example of spontaneous YbIII
reduction. Reaction of [{(Me3Si)2N}3YbIII]with two equiv-
alents of indene gave the expected YbIII product [Eq. (2)].
Reaction with an indene derivative containing chelating
Me2N arms, however, gave exclusively reduction of the
metal [Eq. (3)].[15] Although it is unclear why the presence
of coordinating Lewis bases results in spontaneous reduction
of YbIII, a suitable working hypothesis could be that the YbII
metal in the product has a larger ionic radius, thus allowing for
energetically favorable coordination of additional Lewis
bases (for eight-coordinate species: Yb2+ 1.14, Yb3+
0.985 Š).[16]
[*] C. Ruspic, Prof. Dr. S. Harder
Anorganische Chemie, Universität Duisburg-Essen
Universitätsstrasse 5, 45117 Essen (Germany)
Fax: (+49)201-183-2621
E-mail: sjoerd.harder@uni-due.de
Prof. Dr. J. R. Moss
Department of Chemistry, University of Cape Town
Rondebosch 7701, Cape Town (South Africa)
Dr. M. Schürmann
Anorganische Chemie, Universität Dortmund
44221 Dortmund(Germany)
[**] S.H. kindly acknowledges the University of Cape Town for an
extensive research stay in the autumn of 2003. We are grateful to the
DFG for financial support (SPP1166: “lanthanide-specific func-
tionalities in molecule andmaterial”) andacknowledge Prof. Dr.
Boese andD. Bläser for collection of X-ray data andH. Bandmann
for measurement of 500 MHz NMR spectra.
Encouraged by these results, we investigated the reaction
of [(2-Me2N-benzyl)3Sm]with Cp BIGH. In this case as well, a
reduction of SmIII is observed. The samarocene [(CpBIG)2Sm],
with the metal in its highly reactive 2 + oxidation state, was
Angew. Chem. Int. Ed. 2008, 47, 2121 –2126
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
2121

Communications
organic chemistry,[17] and [Cp*2Sm]is
even able to reduce molecular nitro-
gen.[18] In this light, the spontaneous
reduction of a SmIII precursor to the
samarocene [(CpBIG)2SmII]is remark-
able.
The exact mechanism for this
reduction is unclear. In analogy to the
reaction with yttrium described above
(Scheme 2), the first step could be
formation of the half-sandwich com-
Scheme 1. Reaction of CpBIGH with a dibenzylcalcium derivative.
plex
[(CpBIG)(2-Me2N-benzyl)2Ln].
This complex could be converted to
[(CpBIG)2(2-Me2N-benzyl)Ln]either
by reaction with CpBIGH or by ligand
exchange. Steric crowding in the latter
intermediate could give spontaneous
reduction
to
the
metallocene
[(CpBIG)2Ln]and
a
2-Me 2N-benzyl
radical. This proposed mechanism is
confirmed by detection of 1,2-di(2-
Me2N-phenyl)ethane as a major side
product in the mother liquors. The
mechanism is also supported by the
well-established sterically induced
reduction chemistry (SIR)[19] that has
been discovered by the pioneering
work of Evans et al. on sterically over-
loaded [Cp*3Sm].[20] The latter com-
plex shows the same redox reactivity
[21]
as [Cp*2Sm](Scheme 3).
The per-
formance of [Cp*3Sm]as a reducing
agent has been explained by oxidation
of the Cp* anion. The spontaneous
reduction of SmIII described herein
might shed light on the mechanism of
Scheme 2. Reactions of CpBIGH with benzyllanthanide derivatives.
SIR and suggests that instead of inter-
mediate [Cp*2LnIII]+, transient LnII
species could also play a role. In this
respect, Lappert and co-workersꢀ spec-
troscopic and synthetic evidence for LaII complexes should be
noted.[22]
Considering the bulk of the CpBIG ligand, it seems
reasonable to assume that the spontaneous reduction of
SmIII in [(CpBIG)2(2-Me2N-benzyl)SmIII]is driven by steric
3
=
congestion. On the other hand, [(h -Ph5C5)2W O], a complex
with a much smaller metal, has been characterized by X-ray
diffraction.[23] Therefore, the driving force for the lanthanide
reduction in Scheme 2 could also be partly due to the higher
basicity and thus easier oxidation of the benzyl anion
compared to the Cp* anion. In this respect, it should be
mentioned that a recent new preparative method for
[Cp*3Sm]is based on the reverse of SIR, that is, oxidation
of [Cp*2Sm]by reduction of Cp* Cp* [Eq. (4)].[24] This result
ꢀ
implies that the first step in SIR of [Cp*3Sm]is an equilibrium
that is largely on the side of [Cp*3Sm].
isolated in the form of red-brown crystals (36% yield). The
instability of SmII against oxidation is well-known (Sm3+/Sm2+
E
benzene
80 ꢁC
II
2 ½Cp 3SmIIIŠ ð84 %Þ
1/2 = ꢀ1.55 V).[12] SmI2 is used as a versatile reducing agent in
*
*
*
*
ð4Þ
2 ½Cp 2Sm Š þ Cp ꢀCp
ƒƒƒƒ!
2122
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 2121 –2126

Angewandte
Chemie
refined. These contacts are even shorter when more
realistic (neutron-diffraction-like) calculated hydro-
ꢀ
gen positions with a fixed C H distance of 1.08 Š are
used (Table 1).[30] The observed H···C contacts are of
ꢀ
ꢀ
comparable length to C H···C(Cp ) interactions in
salts [R4P+][Cpꢀ]containing the “naked” Cp
anion.[31e] Concomitant large C H···C angles
ꢀ
[21]
=
(Table 1) indicate an energetically favorable interac-
tion. Such nonclassical hydrogen bonds, arising from
Scheme 3. Reduction of Ph3P S by either [Cp*3Sm] (SIR) or [Cp*2Sm].
Another important driving force for spontaneous LnIII
reduction could be the remarkable stability of the metal-
locene complex [(CpBIG)2Ln]itself. Although formation of
these sterically congested complexes might not seem favor-
able at first sight, more detailed investigations of their crystal
structures shed light on their unusual stability.
The metallocenes [(CpBIG)2M](M = Ca, Yb, Sm) crystal-
¯
lize isomorphously in space group P1 with the metal ion on
the center of inversion (Figure 1). Considering the presence
of many n-butyl substituents, the structures are surprisingly
ordered and of good quality. The propeller-like ligands have
opposite chirality and interlock like a gear box. The metals
are bound in nearly perfect h5,h5 fashion (Table 1). However,
it should be mentioned that for the heavier metals unusually
high displacement factors parallel to the ring planes are
observed (Figure 1b). This anisotropy, which is not as distinct
in the calcocene, is especially noteworthy in the samarocene
(despite measurement at a lower temperature). As heavier
alkaline-earth and LnII metallocenes typically exhibit unusu-
ally bent structures,[25] this anisotropy might be explained by
assuming that the metals are slightly disordered within a plane
parallel to the Cp rings. Consequently, a structure with
parallel ligands but with a bent Cpcenter-M-Cpcenter unit results
(Scheme 4). This implies that there is some deviation from h5
ꢀ
coordination. Parallel ligands but bent C-M C bonds have
also been observed in [{(Me3Si)3C}2Ca].[26]
The average M C bonds in [(CpBIG)2M]
ꢀ
(Table 1) compare remarkably well to those
in the analogous [Cp*2M]complexes (Ca
2.64(1),[25b]
Yb
2.665(3),[25f]
Sm
2.79(1) Š[25g]). As the latter metallocenes
are bent and the Cp* ligand differs substan-
tially from the Ar5C5 ligand,[27] this agree-
ment is merely coincidental. Comparison of
Figure 1. a) Crystal structure of [(CpBIG)2Ca] (ORTEP with thermal
ellipsoids set at 30% probability). Hydrogen atoms have been omitted
for clarity. b) ORTEP representations (top andside views with thermal
ellipsoids set at 50% probability) for partial structures of [(CpBIG)2Ca]
(ꢀ708C), [(CpBIG)2Yb] (ꢀ708C), and[(Cp BIG)2Sm] (ꢀ1008C).
Scheme 4.
ꢀ
the average Yb C bond length in
[(CpBIG)2Yb](2.673(2) Š) with that in the
only other Ar5C5 complex of a lanthanide or alkaline-earth
5
[28]
ꢂ
metal, [{(h -Ph5C5)Yb(m-C CPh)(thf)}2](2.726(7) Š),
led
us to conclude that the Cp M bond in [(CpBIG)2Yb]is
interaction of a Cdꢀ
H
dipole with the aryl p system, are
d+
ꢀ
ꢀ
unexpectedly short.[29]
well-established[31] and explain the typical “herringbone”
structure in crystalline benzene (Scheme 5).[32] The perpen-
One feature in particular could contribute to the unusual
stability of [(CpBIG)2M]complexes. In all three structures,
short distances between an ortho H atom of one ligand and an
ortho C atom of the other ligand are evident (Figure 2). These
merry-go-round interligand contacts are in all [(CpBIG)2M]
complexes considerably shorter than the sum of the van der
ꢀ
Waals radii for C and H (2.90 Š), thus indicating C H···C(p)
hydrogen bonds. Average values are 2.68(1) (M = Ca), 2.74(2)
(M = Yb), and 2.78(2) Š (M = Sm); aryl H atoms have been
Scheme 5.
Angew. Chem. Int. Ed. 2008, 47, 2121 –2126
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2123

Communications
Table 1: Geometrical data for centrosymmetric metallocenes with Ph5C5 or CpBIG ligands (sorted in order
of increasing metal size).[a]
As expected, the interligand
H···C distances in [(Ar5C5)2M]com-
plexes increase with increasing
metal size. However, this increase
is much smaller than the increase in
the ring–ring distance (which equals
[c]
ꢀ
ꢀ
ꢀ
ꢀ
C H···C
Complex
M Cpcenter M C
Cipso CCp ?Cp[b]
H···C[c]
[(Ph5C5)2Fe][6]
1.724(8) 2.04(2)–2.19(2)
+7.8(11)/+13.9(18)
[+11.4(13)]
+3.8(4)/+12.0(4)
[+8.4(13)]
+5.9(4)/+8.9(4)
[+7.2(4)]
+5.1(4)/+7.5(4)
[+6.2(4)]
2.34–2.50 162–177
[2.12(2)]
2.188(5)–2.229(5)
[2.204(5)]
2.268(5)–2.302(5)
[2.281(5)]
2.345(6)–2.371(6)
[2.357(6)]
2.643(1)–2.661(1)
[2.651(1)]
2.666(2)–2.679(2)
[2.673(2)]
2.686(7)–2.705(7)
[2.691(7)]
2.779(2)–2.791(2)
[2.782(2)]
[2.41] [173]
2.39–2.47 144–177
[2.43] [167]
2.41–2.51 149–179
[2.45] [168]
2.45–2.53 169–175
[2.49] [171]
2.51–2.65 153–159
[2.58] [155]
2.54–2.67 152–158
[2.61] [154]
2.63–2.77 150–164
[2.68] [160]
2.61–2.74 152–158
[2.67] [154]
[(Ph5C5)2Ni]+[35] 1.839(2)
[(Ph5C5)2Cr]+[36] 1.932(3)
[(Ph5C5)2W]+[37] 2.018(2)
ꢀ
circa twice the M Cpcenter distance).
This difference originates from
bending of the aryl substituents
out of the Cp plane (Scheme 7).
The
[(Ph5C5)2Fe]strongly bend away
from the metal (average
phenyl
substituents
in
[(CpBIG)2Ca]
[(CpBIG)2Yb]
[(Ph5C5)2Sn][2]
[(CpBIG)2Sm]
2.3561(4)
2.382(1)
2.401(3)
2.5050(8)
+0.031(6)/ꢀ2.31(6)
[ꢀ1.21(6)]
+0.03(1)/ꢀ2.7(1)
[ꢀ1.4(1)]
+ 11.4(13)8). This bending angle
decreases with increasing metal
size. The aryl substituents in
[(CpBIG)2Sm]even bend towards
the metal (average ꢀ3.5(1)8). This
atypical distortion is additional
ꢀ
+0.6(4)/ꢀ2.6(4)
[ꢀ1.2(4)]
ꢀ2.0(1)/ꢀ4.1(1)
[ꢀ3.5(1)]
ꢀ
[a] Distances are given in Š, angles in degrees, and average values between square brackets. [b] Cipso
ꢀ
CCp ?Cp represents the angle between the Cipso CCp vector andthe least-squares plane of the Cp ring ( ꢃ
proof for these attractive
C
values define bending away/towards the metal, see Scheme 7). [c] H···C represents the shortest contact
between the aryl rings of one ligandandthose of the other ligand(see Figure 2). For better comparison
H···C(p) interactions. Although the
energy involved in the H···C contact
is estimated to be 2–5 kcalmolꢀ1,[31]
the overall effect could be consid-
erable. First of all, the molecule
ꢀ
of all structures, the hydrogen atoms have been set at realistic positions with C H bonds of 1.08 Š.
contains ten such contacts, and secondly, cooperative
effects[34] enforce this interaction: the C H hydrogen-bond
donor also functions as the acceptor,
thus increasing acidity and hydrogen-
ꢀ
ꢀ
donor ability of this C H unit.
In summary, contrary to expecta-
tion, [(Ar5C5)2M]complexes are espe-
ꢀ
cially stable on account of a C H···C(p)
hydrogen-bond network that propa-
Scheme 6.
gates through the metallocene. The
high stability of these metallocenes is
expressed by the spontaneous reduction
of a SmIII precursor to the samarocene
[(CpBIG)2Sm]. As this modified metal-
locene and its analogues are readily
soluble in chemically inert solvents such
as hexane and crystallize isomorphously
Figure 2. Partial structure of [(CpBIG)2Ca] showing the merry-go-round
ꢀ
C H···C(p) hydrogen-bond network. Similar structures are observed
for [(CpBIG)2Yb] and[(Cp BIG)2Sm].
for a variety of metal sizes, the CpBIG
Scheme 7.
dicular orientation of the aryl rings in [(CpBIG)2M]complexes
is an S10-symmetric circular version of this herringbone
structure.
ligand will be well-suited for capturing
hitherto unknown highly reducing
metals in low oxidation states. Further exploration of this
intriguing ligand in the chemistry of low-valent metals is in
progress.
ꢀ
Although never discussed earlier, these short C H···C(p)
hydrogen bonds are even more evident in the structures of
[(Ph5C5)2M]complexes with much smaller metals. Entries in
Table 1 are sorted in order of increasing metal size and
ꢀ
contain data for structures in which, for comparison, the C H
Experimental Section
bonds have been set at a value of 1.08 Š. The crystal structure
of [(Ph5C5)2Fe], which unfortunately is of poor quality, shows
extremely short H···C contacts that average 2.41 Š and are of
a similar value to the H···C interaction in the coordination
complex [2-butyne·(HCl)2](Scheme 6). [33] This complex, with
a very acidic hydrogen-bond donor (HCl) and an electron-
All experiments were carried out under argon using standard Schlenk
techniques and freshly dried, degassed solvents. The following
compounds have been prepared according to the literature: (nBu-
C6H4)5CpH (= CpBIGH),[10] [(4-tBu-C6H4CH2)2(thf)4Ca],[38] K[2-
Me2N-benzyl][39] and [(2-Me2N-benzyl)3Y].[40]
[(CpBIG)2Ca]: A solution of [(4-tBu-C6H4CH2)2(thf)4Ca](256 mg,
0.393 mmol) and CpBIGH (285 mg, 0.393 mmol) in benzene (3 mL)
was stirred for 30 min at 608C. All solvents were removed in vacuo
and hexane (3 mL) was added. Slow cooling to ꢀ308C gave large,
ꢀ
rich acceptor (alkyne), is generally seen as the extreme in X
H···C bonding.
2124
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 2121 –2126

Angewandte
Chemie
light-yellow, blocklike crystals of [(CpBIG)2Ca](126 mg, 43% yield
based on CpBIGH). Elemental analysis (%) calcd for C110H130Ca (Mr =
1492.34): C 88.53, H 8.78; found C 88.08, H 8.62. 1H NMR (300 MHz,
C6D6, 208C): d = 0.89 (t, 3J(H,H) = 7.2 Hz, 30H, CH3), 1.28 (m, 20H,
CH2), 1.53 (m, 20H, CH2), 2.46 (t, 3J(H,H) = 7.2 Hz, 20H, CH2), 6.83
(d, 3J(H,H) = 7.6 Hz, 20H, aryl), 7.13 ppm (d, 3J(H,H) = 7.6 Hz, 20H,
aryl). 13C NMR (75 MHz, C6D6, 208C): d = 14.2 (CH3), 22.8 (CH2),
34.2 (CH2), 35.7 (CH2), 124.0, 128.5, 131.9, 134.2, 140.0 ppm. M.p.
2178C.
C 82.02, H 8.31. 1H NMR (300 MHz, C6D6, 208C): d = 2.18 (t,
3J(H,H) = 7.4 Hz, 30H, CH3), 3.06 (m, 20H, CH2), 3.88 (m, 20H,
aryl), 5.13 (t, 3J(H,H) = 7.4 Hz, 20H, CH2), 11.26 (br, 20H, aryl),
15.67 ppm (br, 20H, aryl). 13C NMR (75 MHz, C6D6, 208C): 15.4, 24.5,
37.4, 38.0, 129.8, 137.1, 145.8, 165.0, 183.5 ppm. M.p. 1938C.
Crystal structure determinations: CCDC 665303, 665304, 665305,
665306, and 665307 contain the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif. For all structures: aryl hydrogen atoms were located
and refined isotropically whereas butyl hydrogen atoms were placed
on calculated positions and refined in a riding mode.
[(CpBIG)(2-Me2N-benzyl)2Y]: [(2-Me2N-benzyl)3Y](22.3 mg,
45.4 mmol) and CpBIGH (33.0 mg, 45.4 mmol) were dissolved in
benzene (0.50 mL). After heating for 18 h at 608C, the volatiles
were removed in vacuo (258C, 1 Torr, 30 min). Addition of pentane
(0.40 mL) to the oily residue gave crystallization of the product in the
form of small colorless needles. The crystals were washed with
pentane (2 ” 0.40 mL). Drying under vacuo (258C, 1 Torr, 30 min)
gave the product as a light yellow solid (15.0 mg, 31%). 1H NMR
(300 MHz, C6D6, 208C): d = 0.83 (t, 3J(H,H) = 7.4 Hz, 15H, CH3),
1.20 (m, 10H, CH2), 1.45 (m, 10H, CH2), 2.02 (s, 4H, CH2), 2.33–2.53
(m, 22H, CH2 and NMe2), 6.74 (m, 2H, aryl), 6.85 (d, 3J(H,H) =
8.0 Hz, 10H, aryl), 6.88–6.99 (m, 4H, aryl), 7.09 (m, 2H, aryl),
7.18 ppm (d, 3J(H,H) = 8.0 Hz, 10H, aryl). 13C NMR (75 MHz,
[C6D6] , 208C): d = 14.1 (CH3), 22.6 (CH2), 33.5 (CH2), 35.5 (CH2),
47.2 (NMe2), 48.3 (d, 1J(13C-89Y) = 34.7 Hz, CH2), 118.4, 120.7, 126.8,
127.0, 127.8, 131.1, 132.4, 134.0, 140.0, 144.9, 145.6 ppm (aryl).
[(2-Me2N-benzyl)3Yb]: A cooled solution (ꢀ508C) of K[2-Me2N-
benzyl](2.08 g, 12.03 mmol) in THF (20 mL) was added to a cooled
suspension (ꢀ508C) of YbCl3 (1.12 g, 4.01 mmol) in THF (30 mL). A
rapid color change from orange to dark blue was observed. The
reaction mixture was allowed to warm to 208C and was stirred for 1 h.
After centrifugation the volatiles were evaporated in vacuo (258C,
1 Torr, 30 min) and the residue was recrystallized from toluene/
hexane (ca. 2:1) as dark blue blocks (600 mg, 26%). The crystal
structure is isomorphous to that of [(2-Me2N-benzyl)3Y].[41] Elemen-
tal analysis (%) calcd for C27H36N3Yb (Mr = 575.65): C 56.34, H 6.30;
found C 55.98, H 6.42. M.p. 1578C (decomp).
[(CpBIG)2Ca]: Measurement at ꢀ708C (MoKa), formula
¯
110H130Ca, triclinic, space group P1, a = 13.0492(5), b = 13.9992(3),
C
c = 14.2958(3) Š, a = 117.808(1), b = 97.666(1), g = 93.433(1)8, V=
2267.0(1) Š3, Z = 1, 1calcd = 1.093 gcmꢀ3
,
m
(MoKa) = 0.116 mmꢀ1
,
205883 measured reflections, 18652 independent reflections (Rint
=
0.043), 14568 reflections observed with I > 2s(I), qmax = 34.28, R =
0.0554, wR2 = 0.1639, GOF = 1.01ꢀ,3567 parameters, min/max residual
electron density ꢀ0.31/ + 0.41 eŠ
.
[(CpBIG)2Yb]: Measurement at ꢀ708C (MoKa), formula
¯
C110H130Yb, triclinic, space group P1, a = 13.0726(17), b =
14.0196(9), c = 14.3134(9) Š, a = 117.630(3), b = 98.022(4), g =
93.349(4)8, V= 2278.6(4) Š3, Z = 1, 1calcd = 1.184 gcmꢀ3, m (MoKa) =
1.073 mmꢀ1, 42543 measured reflections, 10454 independent reflec-
tions (Rint = 0.027), 8935 reflections observed with I > 2s(I), qmax
=
27.68, R = 0.0311, wR2 = 0.0829, GOF = 1.02, 567 parameters, min/
ꢀ3
max residual electron density ꢀ0.79/ + 0.49 eŠ
.
[(CpBIG)2Sm]: measurement at ꢀ1008C (MoKa), formula
¯
110H130Sm, triclinic, space group P1, a = 13.0836(11), b =
C
13.8421(12), c = 14.2398(10) Š, a = 117.488(4), b = 97.117(4), g =
94.389(4)8, V= 2243.8(3) Š3, Z = 1, 1calcd = 1.186 gcmꢀ3, m (MoKa) =
0.702 mmꢀ1, 39262 measured reflections, 10247 independent reflec-
tions (Rint = 0.025), 5006 reflections observed with I > 2s(I), qmax
=
27.58, R = 0.0301, wR2 = 0.0612, GOF = 0.82, 567 parameters, min/
ꢀ3
max residual electron density ꢀ0.87/ + 0.29 eŠ
.
[(2-Me2N-benzyl)3Sm]: K[Me2N-benzyl](642 mg, 3.70 mmol) was
treated with SmBr3 (481 mg, 1.23 mmol) analogously to the reaction
with YbCl3. Crystallization from toluene/hexane (ca. 2:1) resulted in
the formation of the product as dark red blocks (143 mg, 21%). The
crystal structure is isomorphous to that of [(2-Me2N-benzyl)3Y].[41]
Elemental analysis (%) calcd for C27H36N3Sm (Mr = 553.0): C 58.65,
H 6.56; found C 58.27, H 6.72. 1H NMR (300 MHz, C6D6, 208C): d =
ꢀ1.77 (br, 18H, NMe2), 4.58 (br, 3H, aryl), 7.12 (br, 3H, aryl), 7.25
(br, 3H, aryl), 9.80 (br, 3H, aryl), 14.25 ppm (br, 6H, CH2). 13C NMR
(75 MHz, C6D6, 208C): d = 39.4 (NMe2), 44.7 (CH2), 111.9, 122.7,
124.3, 128.7, 131.5, 154.7 ppm. M.p. 1508C (decomp).
Received: October 29, 2007
Published online: February 6, 2008
Keywords: calcium · lanthanides · metallocenes ·
.
redox chemistry · steric hindrance
[1]a) C. Janiak, H. Schumann, Adv. Organomet. Chem. 1991, 33,
291; b) R. H. Lowack, K. P. C. Vollhardt, J. Organomet. Chem.
1994, 476, 25.
[(CpBIG)2Yb]: [(2-Me2N-benzyl)3Yb](13.1 mg, 22.7 mmol) and
CpBIGH (33.0 mg, 45.4 mmol) were dissolved in benzene (0.50 mL)
and heated to 608C for 18 h. The volatiles were evaporated in vacuo
(258C, 1 Torr, 30 min) and the residue was dissolved in warm hexane
(0.25 mL). Slow cooling to ꢀ278C resulted in the formation of the
product as large green blocks (25.0 mg, 36%). Elemental analysis (%)
calcd for C110H130Yb (Mr = 1625.30): C 81.29, H 8.06; found C 80.89, H
8.24. 1H NMR (300 MHz, C6D6, 208C): d = 0.89 (t, 3J(H,H) = 7.2 Hz,
30H, CH3), 1.28 (m, 20H, CH2), 1.53 (m, 20H, CH2), 2.46 (t,
[2]M. J. Heeg, C. Janiak, J. J. Zuckerman, J. Am. Chem. Soc. 1984,
106, 4259.
[3]A. Schott, H. Schott, G. Wilke, J. Brandt, H. Hoberg, E. G.
Hoffmann, Justus Liebigs Ann. Chem. 1973, 508.
[4]K. N. Brown, L. D. Field, P. A. Lay, C. M. Lindell, A. F. Masters,
J. Chem. Soc. Chem. Commun. 1990, 408.
[5]L. D. Field, T. W. Hambley, P. A. Humphrey, A. F. Masters, P.
Turner, Inorg. Chem. 2002, 41, 4618.
[6]H. Schumann, A. Lentz, R. Weimann, J. Pickardt, Angew. Chem.
1994, 106, 1827; Angew. Chem. Int. Ed. Engl. 1994, 33, 1731.
[7]J. C. Ruble, H. A. Latham, G. C. Fu, J. Am. Chem. Soc. 1997, 119,
1492.
[8]H. Mahomed, A. Bollmann, J. T. Dixon, V. Gokul, L. Griesel, C.
Grove, F. Hess, H. Maumela, L. Pepler, Appl. Catal. A 2003, 255,
355.
[9]a) S. Harder, F. Feil, K. Knoll, Angew. Chem. 2001, 113, 4391;
Angew. Chem. Int. Ed. 2001, 40, 4261; b) S. Arndt, J. Okuda,
Chem. Rev. 2002, 102, 1953.
3
3J(H,H) = 7.2 Hz, 20H, CH2), 6.83 (d, J(H,H) = 7.6 Hz, 20H, aryl),
7.13 ppm (d, 3J(H,H) = 7.6 Hz, 20H, aryl). 13C NMR (75 MHz, C6D6,
208C): 14.2 (CH3), 22.8 (CH2), 34.2 (CH2), 35.7 (CH2), 124.0, 128.5,
131.9, 134.2, 140.0 ppm. M.p. 1938C.
[(CpBIG)2Sm]: [(2-Me2N-benzyl)3Sm](24.0 mg, 43.4 mmol) and
CpBIGH (63.1 mg, 86.8 mmol) were dissolved in benzene (0.50 mL)
and heated to 608C for 22 h. The volatiles were evaporated in vacuo
(258C, 1 Torr, 30 min). The residue was dissolved in hexane
(0.25 mL). The product crystallized over night at room temperature
in the form of dark brown blocks (25.0 mg, 36%). Elemental analysis
(%) calcd for C110H130Sm (Mr = 1602.62): C 82.44, H 8.18; found
[10]Original preparation: a) H. Schumann, A. Lentz, Z. Naturforsch.
B 1994, 49, 1717; b) General one-pot procedure for Ar5C5H: G.
Angew. Chem. Int. Ed. 2008, 47, 2121 –2126
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
2125

Communications
Dyker, J. Heiermann, M. Miura, J. I. Inoh, S. Pivsa-Art, T. Satoh,
M. Nomura, Chem. Eur. J. 2000, 6, 3426.
[28]C. M. Forsyth, G. B. Deacon, L. D. Field, C. Jones, P. C. Junk,
D. L. Kay, A. F. Masters, A. F. Richards, Chem. Commun. 2006,
1003.
[29]After submission of this manuscript, we presented these results
on the Terrae Rarae in Bonn (November 29, 2007) and learned
[11]In contrast, reaction of [Cp* 2Ca]and [(2-Me 2N-a-Me3Si-
benzyl)2(thf)2Ca]in benzene gave exclusively the heteroleptic
complex [(Cp*)(2-Me2N-a-Me3Si-benzyl)(thf)Ca].
[12]L. R. Morss, Chem. Rev. 1976, 76, 827.
[13]a) T. D. Tilley, R. A. Andersen, B. Spencer, H. Ruben, A. Zalkin,
D. H. Templeton, Inorg. Chem. 1980, 19, 2999; b) L. Arnaudet,
B. Ban, New J. Chem. 1988, 12, 201; c) H. Sitzmann, T.
Dezember, O. Schmitt, F. Weber, G. Wolmershäuser, Z. Anorg.
Allg. Chem. 2000, 626, 2241.
[14]H. A. Zinnen, J. J. Pluth, W. J. Evans, J. Chem. Soc. Chem.
Commun. 1980, 810.
[15]E. Sheng, S. Zhou, S. Wang, G. Yang, Y. Wu, Y. Feng, L. Mao, Z.
Huang, Eur. J. Inorg. Chem. 2004, 2923.
that
G. B.
Deacon
crystallographically
characterized
ꢀ
[(Ph5C5)2Yb]. The Cp Yb distance in this metallocene is similar
to that in [(CpBIG)2Yb].
[30]In crystal structures determined with X-ray diffraction, observed
and refined hydrogen positions always show relatively large
errors and are systematically too short. For comparison, hydro-
gen atoms have been placed at 1.08 Š from the C nucleus (the
ꢀ
ꢀ
average C H bond length) at idealized angles. The C H bond
ꢀ
lengths are hardly influenced by the C H···C(p) interaction. See
also reference [31e].
[31]a) R. Hunter, R. H. Haueisen, A. Irving, Angew. Chem. 1994,
106, 588; Angew. Chem. Int. Ed. Engl. 1994, 33, 566; b) T.
Steiner, J. Chem. Soc. Chem. Commun. 1995, 95; c) T. Steiner,
M. Tamm, A. Grzegorzewski, N. Schulte, N. Veldman, A. M. M.
Schreurs, J. A. Kanters, J. Kroon, J. van der Maas, B. Lutz, J.
Chem. Soc. Perkin Trans. 2 1996, 2441; d) G. R. Desiraju, Chem.
Commun. 2005, 2995; e) S. Harder, Chem. Eur. J. 1999, 5, 1852.
[32]a) G. Karlstrꢁm, P. Linse, A. Wallquist, B. Jꢁnsson, J. Am. Chem.
Soc. 1990, 112, 4768; b) J. M. Steed, F. A. Dixon, W. Klemperer,
J. Chem. Phys. 1979, 70, 4940; c) G. J. Piermarini, A. D. Mighell,
C. E. Weir, S. Block, Science 1969, 165, 1250.
[33]D. Mootz, A. Deeg, J. Am. Chem. Soc. 1992, 114, 5887.
[34]a) A. Karpfen, Adv. Chem. Phys. 2002, 123, 469; b) G. R.
Desiraju, T. Steiner, The Weak Hydrogen Bond, Oxford
University Press, New York, 1997.
[35]L. D. Field, T. W. Hambley, T. He, P. A. Humphrey, C. M.
Lindall, A. F. Masters, Aust. J. Chem. 1996, 49, 889.
[36]L. D. Field, T. W. Hambley, T. He, A. F. Masters, P. Turner, Aust.
J. Chem. 1997, 50, 1035.
[16]R. D. Shannon, Acta Crystallogr. Sect. A 1976, 32, 751.
[17]a) H. B. Kagan, J. Collin, J. L. Namy, C. Bied, F. Dallemer, A.
Lebrun, J. Alloys Compd. 1993, 192, 191; b) V. Nair, A. Deepthi,
Chem. Rev. 2007, 107, 1862.
[18]W. J. Evans, T. A. Ulibarri, J. W. Ziller, J. Am. Chem. Soc. 1988,
110, 6877.
[19]W. J. Evans, Inorg. Chem. 2007, 46, 3435.
[20]W. J. Evans, S. L. Gonzales, J. W. Ziller, J. Am. Chem. Soc. 1991,
113, 1423.
[21]W. J. Evans, K. J. Forestal, J. W. Ziller, J. Am. Chem. Soc. 1998,
120, 9273.
[22]M. C. Cassani, Y. K. Gunꢀko, P. B. Hitchcock, A. G. Hulkes,
A. V. Khvostov, M. F. Lappert, A. V. Protchenko, J. Organomet.
Chem. 2002, 647, 71.
[23]W.-Y. Yeh, C.-L. Ho, M. Y. Chiang, I.-T. Chen, Organometallics
1997, 16, 2698.
[24]W. J. Evans, J. M. Perotti, S. A. Kozimor, T. M. Champagne, B. L.
Davis, G. W. Nyce, C. H. Fujimoto, R. D. Clark, M. A. Johnston,
J. W. Ziller, Organometallics 2005, 24, 3916.
[25]a) R. A. Andersen, J. M. Boncella, C. J. Burns, R. Blom, A.
Haaland, H. V. Volden, J. Organomet. Chem. 1986, 312, C49;
b) R. A. Williams, T. P. Hanusa, J. C. Huffman, Organometallics
1990, 9, 1128; c) M. Kaupp, P. von R. Schleyer, M. Dolg, H. Stoll,
J. Am. Chem. Soc. 1992, 114, 8202; d) T. K. Hollis, J. K. Burdett,
B. Bosnich, Organometallics 1993, 12, 3385; e) T. P. Hanusa,
Organometallics 2002, 21, 2559; f) M. Schultz, C. J. Burns, D. J.
Schwartz, R. A. Andersen, Organometallics 2002, 21, 2559;
g) W. J. Evans, L. A. Hughes, T. P. Hanusa, J. Am. Chem. Soc.
1984, 106, 4270.
[26]C. Eaborn, S. A. Hawkes, P. B. Hitchcock, J. D. Smith, Chem.
Commun. 1997, 1961.
[27]The electronic effects in the Cp* and Ph 5C5 ligands are opposite:
the latter ligand is able to delocalize negative charge in the Ph
rings.
[37]W.-Y. Yeh, S.-M. Peng, G.-H. Lee, J. Organomet. Chem. 1999,
572, 125.
[38]S. Harder, S. Mꢂller, E. Hꢂbner, Organometallics 2004, 23, 178.
[39]S. Harder, F. Feil, Organometallics 2002, 21, 2268.
[40]S. Harder, Organometallics 2005, 24, 373.
[41]The crystal structures of [(2-Me 2N-benzyl)3Yb]and [(2-Me N-
2
benzyl)3Sm]have been determined and will be discussed in a
forthcoming paper. Crystal data for [(2-Me2N-benzyl)3Yb]:
measurement at ꢀ708C (MoKa), formula C27H36N3Yb, mono-
clinic, space group P21/c, a = 16.8969(5), b = 9.4642(3), c =
16.6694(5) Š, b = 111.297(2)8, V= 2483.7(1) Š3, Z = 4, 1calcd
=
1.539 gcmꢀ3. Crystal data for [(2-Me2N-benzyl)3Sm]: measure-
ment at ꢀ1008C (MoKa), formula C27H36N3Sm, monoclinic,
space group P21/c, a = 16.8989(8), b = 9.4648(7), c =
16.6603(10) Š, b = 110.911(3)8, V= 2489.2(3) Š3, Z = 4, 1calcd
=
1.475 gcmꢀ3
.
2126
www.angewandte.org
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2008, 47, 2121 –2126