Mixed-metal lanthanide-iron triple-decker complexes with a cyclo-P 5 building block

Article abstract of DOI:10.1002/anie.201102748

Triple-decker complexes with mixed d/f-block metals that have a purely inorganic middle deck were unknown. A mixed-metal iron-samarium complex is presented with a cyclo-P₅ polyphosphide between the metal centers (see picture; Sm large black, C small black, Fe blue, N green, O red, P pink). This compound can be obtained as either a monomer or a dimer depending on the reaction conditions.

Full text of DOI:10.1002/anie.201102748

DOI: 10.1002/anie.201102748
Polyphosphides
Mixed-Metal Lanthanide–Iron Triple-Decker Complexes with a
cyclo-P5 Building Block**
Tianshu Li, Jelena Wiecko, Nikolay A. Pushkarevsky, Michael T. Gamer, Ralf Kçppe,
Sergey N. Konchenko, Manfred Scheer, and Peter W. Roesky*
In memory of Oskar Glemser (1911–2005)
Triple- and multidecker sandwich complexes have been
discussed in the last decades for their unique electrical and
magnetic properties. The organic spacer between the metals
may facilitate intermetallic electronic communication, which
has a high potential for molecular electronics.[1] A number of
one-dimensional organometallic sandwich molecular wires
(SMWs) have been extensively studied. Thus, the multilayer
vanadium–arene (Ar) organometallic complexes [Vn(Ar)m],
which can be produced in a molecular beam by laser
vaporization, are a class of one-dimensional molecular
magnets.[2] Ferrocene-based molecular wires have been
synthesized in the gas phase and characterized by mass
spectroscopy.[3] It was calculated that these compounds have
half-metallic properties with 100% negative spin polarization
near the Fermi level in the ground state.[4] In contrast to this
investigation in the gas phase, studies on related organome-
tallic triple- and multidecker sandwich complexes containing
f-block elements (lanthanides or actinides) in condensed
phase remain rare;[5] studies were mostly on the cyclooctate-
traene ligand and its derivatives. The only rare-earth-element
triple-decker complex with heterocycles is the low-valent
extensively studied because these compounds exhibit tunable
spectroscopic, electronic, and redox properties, and different
extents of intramolecular p–p interactions.[8] Despite these
promising physical properties further investigations on 4f ele-
ments based triple- and multidecker sandwich complexes are
obviously hampered by the limited variety of ligands that
have been attached to the metal centers to date. Based on
these considerations, we present herein mixed d/f-block-metal
triple-decker complexes with a purely inorganic all-phospho-
rus middle deck.
In contrast to d-block chemistry, where purely inorganic
ring systems of Group 15 elements such as P5 and P6,[9] As5,[9c]
and Sb5[10] are well-established, there is no analogy with the f-
block elements to date. On the other hand, it was shown only
recently that rare-earth elements can stabilize highly reactive
3À [11]
main-group species such as N2
.
Although some heavier
À
Group 15–lanthanide compounds, such as phosphides (Ln
[12d,13]
[14]
PR2),[12] arsenides (Ln AsR2),
stibides (Ln Sb3), and
À
À
bismutides (Ln Bi Bi Ln)[15] are known, the first molecular
polyphosphide of the rare-earth elements, [(Cp*2Sm)4P8]
(Cp* = h5-C5Me5), was recently synthesized.[16] The structure
of the complex is very rare and can be described as a realgar-
type P84À ligand trapped in a cage of four samarocenes. As no
triple-decker sandwich complex of the rare-earth elements
with a polyphosphide middle-deck bridging the metal centers
is known, we focused our interest on the cyclo-P5 ligand. The
structure and properties of this ligand are very similar to the
well-known cyclopentadienyl anion (Scheme 1) and could
therefore have many possible coordination modes.
À
À
À
scandium
1,3,5-triphosphabenzene
complex
[{(h5-
P3C2tBu2)Sc}2(m-h6:h6-P3C3tBu3)], which was obtained by co-
condensation of scandium vaporized in an electron beam with
[6]
ꢀ
an excess of the phosphaalkyne tBuC P. Apart from
organometallic compounds, triple- and multidecker sandwich
complexes of the 4f elements consisting of “salen” type Schiff
base ligands,[7] phthalocyanines, and porphyrins have been
[*] Dr. T. Li, Dr. J. Wiecko, Dr. M. T. Gamer, Dr. R. Kçppe,
Prof. Dr. P. W. Roesky
Institut fꢀr Anorganische Chemie
Karlsruher Institut fꢀr Technologie (KIT)
Engesserstrasse 15, Geb. 30.45, 76131 Karlsruhe (Germany)
E-mail: roesky@kit.edu
Scheme 1. Comparison of the cyclo-P5 and cyclopentadienyl ligands.
Dr. N. A. Pushkarevsky, Prof. S. N. Konchenko
Nikolaev Institute of Inorganic Chemistry SB RAS
Novosibirsk (Russia)
Prof. Dr. M. Scheer
Institut fꢀr Anorganische Chemie
Universitꢁt Regensburg (Germany)
The coordination of cyclo-P5 ligand to an iron(II) center
was first discovered by Scherer and Brꢀck;[9a] the structure[17]
and coordination and redox properties of pentaphosphafer-
rocene, [Cp*Fe(h5-P5)], have been studied intensively. Elec-
trochemistry studies by Geiger and Winter[18] suggested that
the redox properties of [Cp*Fe(h5-P5)] were similar to those
of ferrocene, giving one-electron oxidation/reduction steps.
[Cp*Fe(h5-P5)] is an attractive complex ligand owing to the
possibility of using one or more of the lone pairs of electrons
[**] This work was supported by the Deutsche Forschungsgemeinschaft
(DFG), the Russian Foundation for Basic Research (RFBR), and the
Landesstiftung Baden-Wꢀrttemberg GmbH. We thank Dr. Yanhua
Lan for magnetic susceptibility measurements and Sibylle
Schneider for X-ray single-crystal structure measurement.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201102748.
Angew. Chem. Int. Ed. 2011, 50, 9491 –9495
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9491

Communications
on the phosphorus atoms to coordinate additional metal–
ligand fragments to form spherical molecules,[19] or to take
advantage of the aromatic p-electron system to coordinate a
second metal moiety.[20] Thus, several examples of transition-
metal triple-decker complexes containing the h5-bonded
bridging cyclo-P5 building block have been synthesized in
which the cyclo-P5 unit maintains its pentagonal-planar
arrangement, even under harsh reaction conditions.[9b,21] To
the best of our knowledge, neither rare-earth metal cyclo-P5
complexes nor polyphosphides bridging 4f- and 3d-metal
triple-decker complexes have been reported.
The
reaction
of
the
samarium(II)
complex
[(DIP2pyr)SmI(thf)3][22] (DIP2pyr= 2,5-bis{N-(2,6-diisopro-
pylphenyl)iminomethyl}pyrrolyl) and [Cp*Fe(h5-P5)] in tolu-
ene at elevated temperature resulted in a 1:1 mixture of the
first 3d/4f-cyclo-P5 sandwich complex [Cp*FeP5Sm-
(DIP2pyr)]2 (1) and the corresponding dimeric product
[(DIP2pyr)SmI2(thf)]2 (2) (Scheme 2).
Figure 1. Solid-state structure of 1. Hydrogen atoms are omitted for
clarity. Selected bond lengths [ꢂ] or angles [8]: P1–P2 2.2043(10), P1–
P5 2.1859(9), P2–P3 2.2158(10), P3–P4 2.1706(10), P4–P5 2.2216(11),
Sm–P1 2.8658(8), Sm–P2 3.3803(7), Sm–P3 2.9643(7), Sm–P4
3.0184(7), Sm’–P1 2.9920(7), Sm’–P2 2.8933(7), Sm–N1 2.876(2), Sm–
N2 2.338(2), Sm–N3 2.693(2), Fe–P2 2.2230(7), Fe–P3 2.3165(8), Fe–
P4 2.3162(8), Fe–P5 2.2667(8); P1-P2-P3 110.39(4), P2-P3-P4 97.15(4),
P3-P4-P5 100.33(4), P4-P5-P1 108.14(4), P5-P1-P2 80.79(3).
[(DIP2pyr)SmI(thf)3] and [Cp*Fe(h5-P5)] for 15 h at elevated
temperature in THF in the presence of potassium/naphtha-
lene (Scheme 3). The potassium/naphthalene acted as a
reductant and an iodine abstracting reagent. Single crystals
of the dimeric complex 1 could be obtained by recrystalliza-
tion from toluene and pentane. However, recrystallization of
the product from THF and toluene gave a new monomeric
samarium–iron cyclo-P5 complex [Cp*FeP5Sm(DIP2pyr)-
(thf)2] (3; Scheme 3). In complex 3, the two THF solvent
molecules coordinate to the samarium atom and thus block
the binding sites to prevent formation of the dimeric complex
Scheme 2. Synthesis of compounds 1 and 2. The bonding situation is
simplified.
In the reaction, two equivalents of [(DIP2pyr)SmI(thf)3]
each act as a one-electron reducing agent. One of the
resulting samarium(III) ions is coordinated to the cyclo-P5
unit and the other one is isolated as complex 2. During the
redox reaction one iodine atom was transferred between the
two samarium atoms. As a result, the {(DIP2pyr)Sm} subunit,
which is coordinated to the cyclo-P5 ligand, does not bear any
iodine atom, whereas in compound 2 two iodine atoms are
coordinated to the metal center. Compounds 1 and 2 were
obtained as a mixture of crystals, which could be manually
separated by using a microscope and characterized by X-ray
single-crystal diffraction (Figure 1; Supporting Information,
Figure S1).
To study the formation of 1 and 2 and to obtain analyti-
cally pure products of 1 and 2, the two complexes were
prepared independently. Complex 2 was synthesized by the
reaction of SmI3 and (DIP2pyr)K in THF. Compound 1 could
also be prepared by treatment of only one equivalent of
Scheme 3. Synthesis of compounds 1 and 3. The bonding situation is
simplified.
9492 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 9491 –9495

1. Recrystallization of complex 3 is more accessible than that
of 1. Complex 3 is a more favorable and probable product
than 1, and is even formed sometimes in toluene/pentane
recrystallization. Thus, although we obtained compound 1 on
two independent reaction pathways and present a full
characterization, we consider this compound to be poorly
reproducible.
remaining negative charges are delocalized over the other
four phosphorus atoms and the {Cp*Fe} fragment. Formally,
the cyclo-P5 unit binds in a h4-coordination mode to the iron
atom. The iron part of compounds 1 and 3 thus obey the
À
18 valence-electron rule. The P3 P4 bond (2.1706(10) ꢂ (1)
À
and 2.164(2) ꢂ (3)) is slightly shorter than the other P P
bonds (2.1859(9)–2.2216(11) ꢂ (1) and 2.175(2)–2.228(2) ꢂ
To support the suggested formation pathway of com-
pounds 1 and 3, the oxidation states of the samarium atoms
were determined by magnetic measurements (SQUID) and
NIR spectroscopy. A low paramagnetic susceptibility cM-
(290 K) = 3649 ꢁ 10À6 cm3 molÀ1 and magnetic moment
meff (mB) = 2.91 for two samarium metal centers in 1 suggest
that the oxidation state of samarium is + 3. Similarly, the
(3)). P1 is bent out from the P4 plane at an angle of 124.78 (1)
À
and 127.68 (3). The Sm P distances vary within a wide range.
À
Three Sm P distances in 1 and 3 are within the known
bonding distances reported in the literature.[12b,16] They range
from 2.8658(8) to 3.0184(7) ꢂ (1) and 2.847(2) to
3.0703(15) ꢂ (3). Formally, the h3-coordinated ligand binds
with the negative charged phosphorus atom P1, P3, and P4 to
the samarium center. In compound 1, an additional
h2 coordination of the second cyclo-P5 unit is observed as a
result of the dimerization. These two coordination sites are
occupied by THF molecules in the monomeric compound 3.
To confirm the trivalent oxidation state of the samarium
paramagnetic
susceptibility
cM
(290 K) = 1232 ꢁ
10À6 cm3 molÀ1 and magnetic moment meff (mB) = 1.69 of 3 are
comparable to the literature values for Sm3+ complexes.[23]
Sm3+ complexes exhibit a characteristic absorption pattern in
the NIR even in the presence of a strong visible absorption.[24]
À
The obtained spectra for compounds 1–3 show spectral
centers in complexes 1 and 3, the Sm N bonds were
patterns that are comparable to Sm3+/POCl3/ZrCl4
Sm3+/SeOCl2/SnCl4.[26]
and
compared to similar samarium complexes containing the
(DIP2pyr)À ligand. For comparison we choose the divalent
compound [(DIP2pyr)SmI(thf)3][22] and the trivalent com-
pound 2. The distances of the samarium atom to the pyrrolyl
[25]
Complexes 1–3 were characterized by analytical and
spectroscopic methods and the solid-state structures were
analyzed by single-crystal X-ray diffraction (Figure 1,
Figure 2, and Supporting Information, S1). In the solid state,
complexes 1 and 3 form a pseudo triple-decker structure in
which the cyclo-P5 unit forms the central moiety between the
iron and the samarium atom. As a result of the twofold
reduction, the cyclo-P5 unit adopts an envelope conforma-
tion.[27] In the thus-obtained {Cp*FeP5}2À anion, the phospho-
À
nitrogen atom (Sm N2) of compounds 1 (2.338(2) ꢂ) and 3
(2.385(4) ꢂ) are similar to the trivalent compound
(2.331(2) ꢂ) and differ significantly in comparison with
divalent [(DIP2pyr)SmI(thf)3] (2.474(7) ꢂ).[22] Thus, the data
is in agreement with the magnetic measurements and NIR
spectroscopy.
2
To gain a better insight into the bonding in these novel
monomeric triple-decker compounds, we performed quan-
tum-chemical DFT calculations on 3. The calculated structure
of 3 agrees well with its experimentally obtained parameters
(Supporting Information, Table S2), which shows that the use
of a special ECP and basis set for Sm3+ was a good choice
(Supporting Information). In both 3 and the dianion [Cp*Fe-
3À
rus ligand may be considered as a formally cyclo-P5
polyphosphide anion, which can be understood in the
following way: One negative charge is localized on the
phosphorus atom, which is bent out of plane, whereas the
(h5-P5)]2À [28]
the iron atom is connected to the P5 ring in the
,
way that only four instead of five phosphorus atoms are
connected to iron owing to the bending of P1 out of the
À
resulting P4 plane. The Fe P distances are comparable. The
À
shape of the P5 unit is expected for the formal reduction of P5
3À [29]
À
to P5
.
However, the P P distances in these anions differ
significantly from those found in 3, so that the formulation of
P5 as trianion does not appear to be justified. A better (but not
À
completely satisfactory) agreement is found for the P P
distances calculated for [Cp*Fe(h5-P5)]2À. The samarium atom
À
is threefold coordinated to the P5 ligand. The calculated Sm P
distances agree well with the measured data. Based on
population analyses (Supporting Information), compound 3 is
best interpreted as a {Cp*FeP5Sm}+ cation, which is basically
electrostatically connected to the DIP2pyr anion and two
THF molecules.
In summary, the dimeric and the corresponding mono-
meric compounds 1 and 3 have certain hitherto unknown
features. Compounds 1 and 3 can be considered as the first 3d/
4f-metal triple-decker complexes with a polyphosphide
bridging the metal centers. Both compounds are the first f-
block element compounds ligated by the cyclo-P5 unit. In both
Figure 2. Solid-state structure of 3. Hydrogen atoms are omitted for
clarity. Selected bond lengths [ꢂ] or angles [8]: P1–P2 2.180(3), P1–P5
2.175(2), P2–P3 2.228(2), P3–P4 2.164(2), P4–P5 2.215(3), Sm–P1
2.847(2), Sm–P2 3.3385(15), Sm–P3 2.980(2), Sm–P4 3.0703(15), Sm–
N1 2.769(5), Sm–N2 2.385(4), Sm–N3 2.770(5), Sm–O1 2.524(4),
Sm–O2 2.496(4), Fe–P2 2.262(2), Fe–P3 2.291(1), Fe–P4 2.321(1), Fe–
P5 2.255(2); P1-P2-P3 110.27(9), P2-P3-P4 99.32(10), P3-P4-P5
99.21(9), P4-P5-P1 110.08(9), P5-P1-P2 82.82(9).
Angew. Chem. Int. Ed. 2011, 50, 9491 –9495
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9493

Communications
stirred for 16 h at 608C. Both bright yellow crystals of 2 and dark
orange crystals of 1 were obtained after the 2-section ampoule
technique procedure. The crystals could be differentiated and
separated under a microscope.
compounds reduction of the cyclo-P5 unit leads to the
{Cp*FeP5}2À subunit in which the phosphorus ligand may be
3À
considered as a formally cyclo-P5 polyphosphide anion.
Mixed 3d/4f-element triple-decker complexes raise funda-
mental questions about the formation of polyphosphides and
other phosphorus derivatives[30] within the coordination
sphere of a metal center. Furthermore, they open new
possibilities towards a better comprehension of intermetallic
communication.
Crystal data for 1: C80H106Fe2N6P10Sm2·3(C7H8), Mr = 2150.21,
triclinic, a = 13.8374(7), b = 14.7963(10), c = 14.8908(8) ꢂ, a =
65.526(4), b = 70.167(4), g = 68.100(5)8, V= 2512.2(3) ꢂ3, T=
150(2) K, space group P1, Z = 1, m(MoKa) = 1.642 mmÀ1, 18967
¯
reflections measured, 9037 independent reflections (Rint = 0.0353).
The final R1 values were 0.0258 (I > 2s(I)). The final wR(F2) values
were 0.0432 (all data). The goodness of fit on F2 was 0.998.
2: C68H96I4N6O2Sm2·2(C7H8), Mr = 2022.08, triclinic, a =
15.1260(4), b = 16.6343(5), c = 18.9589(6) ꢂ, a = 106.169(3), b =
90.251(3), g = 110.110(2)8, V= 4275.7(2) ꢂ3, T= 200(2) K, space
Experimental Section
group P1, Z = 2, m(MoKa) = 2.849 mmÀ1, 32012 reflections measured,
1: THF (10 mL) was condensed at À788C onto a mixture of
[(DIP2pyr)SmI(thf)3] (185 mg, 0.20 mmol), [Cp*Fe(h5-P5)] (70 mg,
0.20 mmol), potassium metal (8 mg, 0.20 mmol), and naphthalene
(30 mg, 0.24 mmol). The resulting reaction mixture was stirred for
16 h at 608C. The solvent of the dark-colored solution was evaporated
to dryness. Toluene (5 mL) was condensed onto the dark residue to
give a dark brown solution. The solution was filtered off and layered
with pentane. Dark orange crystals and a pale precipitate were
obtained at ambient temperature after three days. Yield: 35 mg, 16%
(single crystals). Magnetic susceptibility: cM (290 K) = 3649 ꢁ
10À6 cm3 molÀ1, meff (mB) = 2.91. MIR(KBr): n˜ = 2958 (m), 2902 (w),
1620 (s), 1587 (m), 1459 (m), 1439 (m), 1376 (m), 1309 (m), 1230 (m),
1155 (vs), 1022 (vs), 860 (m), 783 (s), 728 (s), 497 (w), 464 (s), 439 cmÀ1
(m). NIR(KBr): n˜ = 9276 (6F9/2), 8155 (6F7/2), 7308 (6F5/2,), 6786 (6F3/2),
6479 cmÀ1 (6F1/2). Anal. calcd (%) for Sm2Fe2P10N6C80H106
(1873.91 gmolÀ1): C 51.28, H 5.70, N 4.48; found: C 51.16, H 5.82,
N 4.03.
¯
15115 independent reflections (Rint = 0.0350). The final R1 values
were 0.0241 (I > 2s(I)). The final wR(F2) values were 0.0552 (all
data). The goodness of fit on F2 was 1.003.
3: 2(C48H69FeN3O2P5Sm)·3(C7H8), Mr = 2438.63, triclinic, a =
12.8658(8), b = 13.3175(9), c = 19.6233(13), a = 77.726(5), b =
71.236(5), g = 66.402(5)8, V= 2903.7(3) ꢂ3, T= 150(2) K, space
group P1, Z = 1, m(MoKa) = 1.432 mmÀ1, 22189 reflections measured,
¯
10330 independent reflections (Rint = 0.0844). The final R1 values
were 0.0540 (I > 2s(I)). The final wR(F2) values were 0.1385 (all
data). The goodness of fit on F2 was 1.005.
CCDC 822355 (1), CCDC 822356 > (2), and CCDC 822357(3)
contains 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.
2: THF (10 mL) was condensed at À788C onto a mixture of SmI3
(200 mg, 0.380 mmol) and (DIP2pyr)K (180 mg, 0.375 mmol). The
resulting reaction mixture was stirred for 16 h at 608C. The solvent of
the orange solution was evaporated to dryness. Toluene (5 mL) was
condensed onto the yellow residue to give yellow–orange solution and
a pale precipitate. The mixture was filtered off into a 2-section
ampoule to grow crystals by slow evaporation. Yellow crystals were
obtained at ambient temperature. Yield: 320 mg, 92.7%. 1H NMR
(300 MHz, C6D6): d = 0.60 (brs, 4H, N = CH), 1.08 (d, 48H, J = 7 Hz,
CH(CH3)2), 1.40 (brs, 8H, THF), 3.03 (sept, 8H, J = 7 Hz, CH-
(CH3)2), 3.59 (brs, 8H, THF), 6.37 (s, 4H, 3,4-pyr), 7.01–10.37 ppm
(m, 12H, Ph). NIR(KBr): n˜ = 9211, (6F9/2), 8021 (6F7/2), 7148 (6F5/2,),
6360 (6F1/2), 6246 cmÀ1 (6H15/2.). Anal. calcd (%) for 1·(2C6H5CH3),
Sm2I4N6O2C82H108 (2018.12 gmolÀ1): C 48.80, H 5.39, N 4.16; found:
C 48.54, H 5.68, N 3.917.
Received: April 20, 2011
Revised: June 6, 2011
Published online: August 31, 2011
Keywords: iron · phosphorus · polyphosphides · samarium ·
.
triple-decker complexes
[1] a) D. Astruc, Acc. Chem. Res. 1997, 30, 383 – 391; b) P. Nguyen, P.
Gomez-Elipe, I. Manners, Chem. Rev. 1999, 99, 1515 – 1548.
[2] a) K. Miyajima, A. Nakajima, S. Yabushita, M. B. Knickelbein,
K. Kaya, J. Am. Chem. Soc. 2004, 126, 13202 – 13203; b) J. Wang,
P. H. Acioli, J. Jellinek, J. Am. Chem. Soc. 2005, 127, 2812 – 2813.
[3] S. Nagao, A. Kato, A. Nakajima, K. Kaya, J. Am. Chem. Soc.
2000, 122, 4221 – 4222.
3: THF (10 mL) was condensed at À788C onto a mixture of
[(DIP2pyr)SmI(thf)3] (375 mg, 0.40 mmol), [Cp*Fe(h5-P5)] (140 mg,
0.40 mmol), potassium metal (19 mg, 0.49 mmol) and naphthalene
(55 mg, 0.43 mmol) in a Schlenk tube. The resulting reaction mixture
was stirred for 16 h at 608C. After the solvent was removed, the dark
brown solid was washed with heptanes (20 mL) and dried under
vacuum. THF (10 mL) was condensed onto the solid and stirred for
1 h. The mixture was filtered into a 2-section ampoule. Toluene
(10 mL) was condensed to the filtrate to grow crystals in a 2-section
ampoule. The dark orange crystals were obtained at ambient temper-
ature after three days. Yield: 120 mg, 28% (single crystals). Magnetic
[4] L. Zhou, S.-W. Yang, M.-F. Ng, M. B. Sullivan, Tan, L. Shen, J.
Am. Chem. Soc. 2008, 130, 4023 – 4027.
[5] F. T. Edelmann, New J. Chem. 2011, 35, 517 – 528.
[6] P. L. Arnold, F. G. N. Cloke, P. B. Hitchcock, J. F. Nixon, J. Am.
Chem. Soc. 1996, 118, 7630 – 7631.
[7] X. Yang, R. A. Jones, J. Am. Chem. Soc. 2005, 127, 7686 – 7687.
[8] J. Jiang, D. K. P. Ng, Acc. Chem. Res. 2008, 42, 79 – 88.
[9] a) O. J. Scherer, T. Brꢀck, Angew. Chem. 1987, 99, 59; Angew.
Chem. Int. Ed. Engl. 1987, 26, 59; b) O. J. Scherer, T. Brꢀck, G.
Wolmershꢃuser, Chem. Ber. 1989, 122, 2049 – 2054; c) O. J.
Scherer, H. Sitzmann, G. Wolmershꢃuser, Angew. Chem. 1985,
97, 358 – 359; Angew. Chem. Int. Ed. 1985, 24, 351 – 353.
[10] H. J. Breunig, N. Burford, R. Rçsler, Angew. Chem. 2000, 112,
4320 – 4322; Angew. Chem. Int. Ed. 2000, 39, 4148 – 4150.
[11] M. Fang, D. S. Lee, J. W. Ziller, R. J. Doedens, J. E. Bates, F.
Furche, W. J. Evans, J. Am. Chem. Soc. 2011, 133, 3784 – 3787.
[12] a) H. Schumann, E. Palamidis, G. Schmid, R. Boese, Angew.
Chem. 1986, 98, 726 – 727; Angew. Chem. Int. Ed. 1986, 25, 718 –
719; b) G. W. Rabe, G. P. A. Yap, A. L. Rheingold, Inorg. Chem.
1997, 36, 3212 – 3215; c) M. R. Douglass, C. L. Stern, T. J. Marks,
J. Am. Chem. Soc. 2001, 123, 10221 – 10238; d) W. J. Evans, J. T.
susceptibility cM (290 K) = 1232 ꢁ 10À6 cm3molÀ1
,
meff (mB) = 1.69.
MIR(KBr): n˜ = 2959 (vs), 2900 (m), 1623 (m), 1568 (vs), 1449 (s),
1371 (m), 1333 (s), 1250 (m), 1162 (s), 1099 (m), 1050 (s), 927 (m), 846
(s), 768 (m), 735 (s), 564 (m), 463 cmÀ1 (m). NIR(KBr): n˜ = 9216 (6F9/
2), 7925 (6F7/2), 7218 (6F5/2,), 6631 (6F3/2), 6430 (6F1/2), 6321 cmÀ1 (6H15/
2.). Anal. calcd (%) for SmFeP5N3O2C48H69 (1081.17 gmolÀ1): C 53.32,
H 6.43, N 3.89; found: C 53.71, H 6.32, N 3.83.
Reaction of [(DIP2pyr)SmI (thf)3] and [Cp*Fe(h5-P5)]: Toluene
(10 mL) was condensed at À788C onto a mixture of [(DIP2pyr)SmI-
(thf)3] (108 mg, 0.12 mmol) and [Cp*Fe(h5-P5)] (20 mg, 0.058 mmol)
in one side of a 2-section ampoule. The resulting reaction mixture was
9494 www.angewandte.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2011, 50, 9491 –9495

Leman, J. W. Ziller, S. I. Khan, Inorg. Chem. 1996, 35, 4283 –
4291.
[13] H. Schumann, E. Palamidis, J. Loebel, J. Pickardt, Organo-
metallics 1988, 7, 1008 – 1010.
b) B. Rink, O. J. Scherer, G. Heckmann, G. Wolmershꢃuser,
Chem. Ber. 1992, 125, 1011 – 1016; c) A. R. Kudinov, P. V.
Petrovskii, M. I. Rybinskaya, Russ. Chem. Bull. 1999, 48,
1362 – 1364.
[14] W. J. Evans, S. L. Gonzales, J. W. Ziller, J. Chem. Soc. Chem.
Commun. 1992, 1138 – 1139.
[22] J. Jenter, M. T. Gamer, P. W. Roesky, Organometallics 2010, 29,
4410 – 4413.
[15] W. J. Evans, S. L. Gonzales, J. W. Ziller, J. Am. Chem. Soc. 1991,
113, 9880 – 9882.
[16] S. N. Konchenko, N. A. Pushkarevsky, M. T. Gamer, R. Koeppe,
H. Schnçckel, P. W. Roesky, J. Am. Chem. Soc. 2009, 131, 5740 –
5741.
[17] M. Scheer, L. J. Gregoriades, A. V. Virovets, W. Kunz, R.
Neueder, I. Krossing, Angew. Chem. 2006, 118, 5818 – 5822;
Angew. Chem. Int. Ed. 2006, 45, 5689 – 5693.
[18] a) R. F. Winter, W. E. Geiger, Organometallics 2003, 22, 1948 –
1952; b) R. F. Winter, W. E. Geiger, Organometallics 1999, 18,
1827 – 1833.
[19] a) J. Bai, A. V. Virovets, M. Scheer, Science 2003, 300, 781 – 783;
b) S. Welsch, C. Grçger, M. Sierka, M. Scheer, Angew. Chem.
2011, 123, 1471 – 1474; Angew. Chem. Int. Ed. 2011, 50, 1435 –
1438; c) M. Scheer, A. Schindler, J. Bai, B. P. Johnson, R. Merkle,
R. Winter, A. V. Virovets, E. V. Peresypkina, V. A. Blatov, M.
Sierka, H. Eckert, Chem. Eur. J. 2010, 16, 2092 – 2107; d) M.
Scheer, Dalton Trans. 2008, 4372 – 4386.
[23] F. T. Edelmann, P. Poremba in Synthetic Methods of Organome-
tallic and Inorganic Chemistry, Vol. 6 (Ed.: W. A. Herrmann),
Georg Thieme Verlag, Stuttgart, 1997.
[24] a) H. D. Amberger, C. Hagen, Spectrochim. Acta Part A 1994,
50, 1267 – 1275; b) H.-D. Amberger, H. Reddmann, W. J. Evans,
Inorg. Chem. 2009, 48, 10811 – 10818.
[25] K. Bukietynska, B. Radomska, Polyhedron 1983, 2, 1297 – 1299.
[26] Y. C. Ratnakaram, S. Buddhudu, Mater. Chem. Phys. 1985, 12,
443 – 448.
[27] Other envelope-shaped compounds were obtained by the
reactions of [Cp*Fe(h5-P5)] with the organometallic compounds
[(h5-C5H3tBu2)Ta(CO)4] and [{Cp*Ir(CO)}2]: M. Detzel, T.
Mohr, O. J. Scherer, G. Wolmershꢃuser, Angew. Chem. 1994,
106, 1142 – 1144; Angew. Chem. Int. Ed. 1994, 33, 1110 – 1112.
[28] The ground state of the dianion of [Cp*FeP5] is a singlet.
Geometry optimization of the lowest-lying 3A’’ state (disfavored
by 49 kJmolÀ1) leads to separated Cp* and FeP5À anions. For this
reason, we did not consider this state in the discussion.
3À
[20] M. Caporali, L. Gonsalvi, A. Rossin, M. Peruzzini, Chem. Rev.
2010, 110, 4178 – 4235.
[29] The 1A’ and 3A’’ states of P5 are of comparable energy; the
3A’’ state is disfavored by 12 kJmolÀ1
.
[21] a) A. R. Kudinov, D. A. Loginov, Z. A. Starikova, P. V. Petrovs-
kii, M. Corsini, P. Zanello, Eur. J. Inorg. Chem. 2002, 3018 – 3027;
[30] N. A. Piro, J. S. Figueroa, J. T. McKellar, C. C. Cummins, Science
2006, 313, 1276 – 1279.
Angew. Chem. Int. Ed. 2011, 50, 9491 –9495
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 9495