Versatile new C3-symmetric tripodal tetraphosphine ligands; structural flexibility to stabilize Culand Rhl species and tune their reactivity

Article abstract of DOI:10.1021/ic100221w

The high-yielding synthesis and detailed characterization of two well-defined, linkage isomeric tripodal, tetradentate allphosphorus ligands 1-3 Is described. Coordination to Cu^l resulted in formation of complexes 4-6, for which the molecular structures indicate overall tridentate coordination to the copper atom in the solid state, with one dangling peripheral phosphine. The solution studies suggest fast exchange between the three phosphine side-arms. For these new Cu^l complexes, preliminary catalytic activity In the cyclopropanation of styrene with ethyldiazoacetate (EDA) is disclosed. The anticipated well-defined tetradentate coordination in a C ₃-symmetric fashion was achieved with Rh^l and lr ^l, leading to the overall five-coordinated complexes 7-12. Complex 11 has the norbornadiene (nbd) ligand coordinated in an unprecedented monodentate 2,3-η² mode to Rh. Furthermore, unexpected but very interesting redox-chemistry and reactivity was displayed by the Rh(Cl)-complexes 7 and 8. Oxidation resulted In the formation of stable Rh^ll metalloradicals [7]PF₆ and [8]PF₆ that were characterized by X-ray crystallography, magnetic susceptibility measurements, cyclic voltammetry, and electron paramagnetic resonance (EPR) spectroscopy. Subsequent redox-reactivity of these metalloradicals toward molecular hydrogen is described, resulting in the formation of Rh^lll hydride compounds.

Full text of DOI:10.1021/ic100221w

Inorg. Chem. 2010, 49, 6495–6508 6495
DOI: 10.1021/ic100221w
Versatile New C3-Symmetric Tripodal Tetraphosphine Ligands; Structural
Flexibility to Stabilize CuI and RhI Species and Tune Their Reactivity
Jeroen Wassenaar,† Maxime A. Siegler,‡,§ Anthony L. Spek,‡ Bas de Bruin,† Joost N. H. Reek,*,† and
Jarl Ivar van der Vlugt*,†
†Supramolecular & Homogeneous Catalysis Group, van ‘t Hoff Institute for Molecular Sciences,
University of Amsterdam, Nieuwe Achtergracht 166, 1018 WV Amsterdam, The Netherlands, and
‡Department of Crystal and Structural Chemistry, University of Utrecht, Padualaan 8, 3584 CH Utrecht,
The Netherlands. §Current address: Department of Chemistry, John Hopkins University, Baltimore, Maryland
Received February 3, 2010
The high-yielding synthesis and detailed characterization of two well-defined, linkage isomeric tripodal, tetradentate all-
phosphorus ligands 1-3 is described. Coordination to CuI resulted in formation of complexes 4-6, for which the molecular
structures indicate overall tridentate coordination to the copper atom in the solid state, with one dangling peripheral
phosphine. The solution studies suggest fast exchange between the three phosphine side-arms. For these new CuI
complexes, preliminary catalytic activity in the cyclopropanation of styrene with ethyldiazoacetate (EDA) is disclosed. The
anticipated well-defined tetradentate coordination in a C3-symmetric fashion was achieved with RhI and IrI, leading to the
overall five-coordinated complexes 7-12. Complex 11 has the norbornadiene (nbd) ligand coordinated in an
unprecedented monodentate 2,3-η2 mode to Rh. Furthermore, unexpected but very interesting redox-chemistry and
reactivity was displayed by the Rh(Cl)-complexes 7 and 8. Oxidation resulted in the formation of stable RhII metalloradicals
[7]PF6 and [8]PF6 that were characterized by X-ray crystallography, magnetic susceptibility measurements, cyclic
voltammetry, and electron paramagnetic resonance (EPR) spectroscopy. Subsequent redox-reactivity of these
metalloradicals toward molecular hydrogen is described, resulting in the formation of RhIII hydride compounds.
Introduction
In particular C3-symmetric tripodal skeletons have
emerged as attractive frameworks to address specific signifi-
cant research questions and to steer metal-centered
reactivity.4-6 Within this family of scaffolds, the pivotal
Constrained metal geometries are often a prerequisite for
unusual reactivity and selective metal-mediated catalysis, and
this is prevalent as a design principle in both biological and
synthetic systems.1 Multipodal ligand scaffolds are a common
approach to enforce formation of metal complexes in well-
defined yet unusual and reactive geometries. Notwithstanding
the long history of such sterically encumbered, highly regular,
multidentate scaffolds, Trofimenko’s scorpionates being a
classic example2, their chemistry as well as the development
of new ligand classes continue to receive much attention.3
(4) Monoanionic tris(phosphino)borates:(a) Lu, C. C.; Saouma, C. T.;
Day, M. W.; Peters, J. C. J. Am. Chem. Soc. 2007, 129, 4–5. (b) Betley, T. A.;
Peters, J. C. Inorg. Chem. 2003, 42, 5074–5084. (c) Shapiro, I. R.; Jenkins, D. M.;
Thomas, J. C.; Day, M. W.; Peters, J. C. Chem. Commun. 2001, 2152–2153. (d)
Turculet, L.; Feldman, J .D.; Tilley, T. D. Organometallics 2004, 23, 2488–2502.
Related tris(phosphino)silanes: Mankad, N. P.; Whited, M. T.; Peters, J. C.
Angew. Chem., Int. Ed. 2007, 46, 5768–5771. Whited, M. T.; Mankad, N. P.;
Lee, Y.; Oblad, P. F.; Peters, J. C. Inorg. Chem. 2009, 48, 2507–2517. Lee, Y.;
Mankad, N. P.; Peters, J. C. Nat. Chem. 2010, 2, DOI: 10.1038/nchem.660.
(5) Neutral tris(phosphino)boranes: (a) Sircoglou, M.; Bontemps, S.;
Bouhadir, G.; Saffon, N.; Miqueu, K.; Gu, W. X.; Mercy, M.; Chen,
C. H.; Foxman, B. M.; Maron, L.; Ozerov, O. V.; Bourissou, D. J. Am.
Chem. Soc. 2008, 130, 16729–16738. (b) Bontemps, S.; Bouhadir, G.; Gu, W.;
Mercy, M.; Chen, C. H.; Foxman, B. M.; Maron, L.; Ozerov, O. V.; Bourissou, D.
Angew. Chem., Int. Ed. 2008, 47, 1481–148. (c) Bontemps, S.; Bouhadir, G.;
Dyer, P. W.; Miqueu, K.; Bourissou, D. Inorg. Chem. 2007, 46, 5149–5151. For
other C3-symmetric neutral phosphorus-based ligands, see: (d) Grutters, M. M. P.;
van der Vlugt, J. I.; Pei, Y.; Mills, A. M.; Lutz, M.; Spek, A. L.; M€uller, C.;
Moberg, C.; Vogt, D. Adv. Synth. Catal. 2009, 351, 2199–2208. (e) Ionescu, G.;
van der Vlugt, J. I.; Abbenhuis, H. C. L.; Vogt, D. Tetrahedron: Asymmetry
2006, 16, 3970–3975. (f ) van der Vlugt, J. I.; Ackerstaff, J.; Dijkstra, T. W.; Mills,
A. M.; Kooijman, H.; Spek, A. L.; Meetsma, A.; Abbenhuis, H. C. L.; Vogt, D.
Adv. Synth. Catal. 2004, 346, 399–412.
*To whom correspondence should be addressed. E-mail: j.n.h.reek@
uva.nl (J.N.H.R.), j.i.vandervlugt@uva.nl (J.I.v.d.V.).
(1) (a) van der Vlugt, J. I.; Koblenz, T.; Wassenaar, J.; Reek, J. N. H. In
Molecular Encapsulation: Reactions in Constrained Systems; Mieusset, J.-L.;
Brinker, U. H., Eds.; Wiley-VCH: New York, 2010; (b) van der Vlugt, J. I.; Reek,
J. N. H. Angew. Chem., Int. Ed. 2009, 48, 8832–8846. (c) Koblenz, T.; Wassenaar,
J.; Reek, J. N. H. Chem. Soc. Rev. 2008, 37, 247–262. (d) Kuil, M.; Soltner, T.; van
Leeuwen, P. W. N. M.; Reek, J. N. H. J. Am. Chem. Soc. 2006, 128, 11344–11345.
(e) Drauz, K.; Waldmann, H. Enzyme Catalysis in Organic Synthesis: A
Comprehensive Handbook, 2nd ed.; Wiley-VCH: New York, 2002.
(2) Trofimenko, S. Chem. Rev. 1993, 93, 943–980.
(3) (a) Moberg, C. Angew. Chem., Int. Ed. 1998, 37, 248–268. (b) Kuzu, I.;
Krummenacher, I.; Meyer, J.; Armbruster, F.; Breher, F. Dalton Trans. 2008,
5836–5865.
r
2010 American Chemical Society
Published on Web 06/16/2010
pubs.acs.org/IC

6496 Inorganic Chemistry, Vol. 49, No. 14, 2010
Wassenaar et al.
species, which was successfully applied in homogeneous
catalysis. Excellent results were obtained in the synthesis of
Roche ester derivatives by asymmetric hydrogenation.14 We
have extended this work to chiral phosphole-phosphorami-
dite ligands, for use in Pd-catalyzed asymmetric allylic
alkylation.15
We next wondered if we could expand on this motif
to arrive at sterically demanding tripodal tetraphosphine
ligands (PP3) and to see if the corresponding metal complexes
with such enforcing frameworks display unusual character-
istics. Such scaffolds should enforce conformational rigidity
and in particular unusual geometries around (late) transition
metal centers, potentially leading to new reactivity. Here we
report the synthesis of new C3-symmetric tripodal tetraphos-
phine ligands 1-3, capitalizing on previously developed
modular synthetic strategies for IndolPhos, and their com-
plexes with CuI, RhI, and IrI. Reactivity studies of these
complexes show facile oxidation of the RhI complexes,
resulting in stable RhII metalloradicals. The dominant ligand
structure results in highly distorted square-pyramidal geo-
metries. This in turn is the onset for unusual subsequent
reactivity of these paramagnetic species with dihydrogen to
yield RhIII hydride compounds.
Figure 1. Tetraphosphine A and the new indole-based all-phosphorus
tripodal ligands 1-3.
atom can either be a spectator entity (tridentate ligation) or a
donor atom binding to the metal fragment (tetradentate
ligation). C3-symmetric tripodal, tetradentate ligands that
feature an all-phosphorus coordination sphere are rare.7,8
Bianchini et al. studied the coordination chemistry of ligand
A toward Rh,9,10 while Field and co-workers recently de-
scribed Fe(0) and Ru(0) dinitrogen complexes with the iso-
propyl-derivative of ligand A (see Figure 1).11
The applicability of such C3 platforms for the formation of
(late) transition metal complexes is related to the confined
geometry dictated onto the metal center. Subtle electronic
influences, for example, of the pivotal atom and the side arm
donor groups, and the steric impact of peripheral fragments
govern the resulting reactivity of (in particular) low-valent
metal species.
Results and Discussion
We set out to construct PP3 ligand 1, which features
a central phosphorus atom connected to three indole
N-fragments.16 This ligand is straightforwardly synthesized
from the reported 2-diphenylphosphino-3-methylindole via a
Convergent synthetic methodologies to prepare symmetric
tripodal scaffolds are available but most of the reported
examples lack opportunities to easily access a number of
electronic and structural variations through modular synth-
esis. Moreover, despite the highly symmetric nature of the
C3-symmetric backbone, rigidity is scarcely employed as an
important design feature for tripodal ligand systems.12 How-
ever, flexidentate behavior13 of the scaffold is undesirable
when targeting restricted geometric structures to study the
effects on corresponding metal centers and their reactivity.
We recently reported on the use of 3-methylindole as rigid
scaffold for the construction of hybrid bidentate ligand
IndolPhos (Figure 2), a chiral phosphine-phosphoramidite
base-promoted reaction with PCl3. In the corresponding 31
P
NMR spectrum, the signal for the pivotal P-atom was
observed as a quartet at δ 73.7 ppm, while a doublet was
observed at δ -29.6 ppm for the peripheral PPh2 units, with a
large coupling constant JP-P of 197 Hz. To demonstrate the
versatile nature of these scaffolds, we also prepared and fully
characterized the tris-PiPr2 analogue 2 of compound 1. It
shows two distinct peaks in the 31P NMR spectrum at δ -9.9
(PP3) and 67.3 ppm (PP3) with the expected relative integra-
tion ratio of 1:3. Because of the rigid, sterically encumbered
conformation of ligand 2, the two isopropyl groups on each
peripheral phosphine fragment are inequivalent, even in
(6) Monoanionic tris(indolylphosphino)methane: (a) Ciclosi, M.; Estevan,
ꢀ
F.; Lahuerta, P.; Passarelli, V.; Perez-Prieto, J.; Sanau, M. Dalton Trans. 2009,
i
solution, as indicated by the observation of two Pr -CH
ꢀ
2290–2297. (b) Ciclosi, M.; Estevan, F.; Lahuerta, P.; Passarelli, V.; Perez-Prieto, J.;
group resonances in the 1H NMR spectrum. We ascribe this
observation to a locked inequivalent conformation of the iPr
groups, caused by strong steric interactions between the
neighboring P(iPr)2 moieties. Four different iPr-based methyl
signals in a 1:1:1:1 ratio are expected, based on this inequi-
valency of the two iPr groups per P(iPr)2 moiety, combined
with the diastereotopicity of the two Me groups within each
iPr moiety and the overall C3 symmetry of the ligand (leading
to three equivalent P(iPr)2 moieties). Two of these signals
apparently overlap, since we observe three iPr methyl signals
Sanau, M. Adv. Synth. Catal. 2008, 350, 234–236. (c) Ciclosi, M.; Lloret, J.;
ꢀ
Estevan, F.; Lahuerta, P.; Sanau, M.; Perez-Prieto, J. Angew. Chem., Int. Ed. 2006,
45, 6741–6744. For a report onthe samescaffold but lackingthe relevant pivotal C-H
ꢀ
activation: Ciclosi, M.; Lloret, J.; Estevan, F.; Sanaꢀu, M.; Perez-Prieto, J. Dalton
Trans. 2009, 5077.
(7) Baker, M. J.; Pringle, P. G. J. Chem. Soc., Chem. Commun. 1993, 314–
316.
(8) King, R. B.; Kapoor, R. N. J. Am. Chem. Soc. 1969, 91, 5191–5192.
(9) (a) Bianchini, C.; Masi, D.; Meli, A.; Peruzzini, M.; Zanobini, F.
J. Am. Chem. Soc. 1988, 110, 6411–6423. (b) Bianchini, C.; Meli, A.; Peruzzini,
M.; Zanobini, F. Organometallics 1990, 9, 1155–1160. (c) Bianchini, C.; Mealli,
C.; Peruzzini, M.; Zanobini, F. J. Am. Chem. Soc. 1987, 109, 5548–5549.
(d) Bianchini, C.; Meli, A.; Peruzzini, M.; Vacca, A.; Zanobini, F. Organome-
tallics 1987, 6, 2453–2455.
(14) (a) Wassenaar, J.; Reek, J. N. H. Dalton Trans. 2007, 3750–3753. (b)
Wassenaar, J.; Kuil, M.; Reek, J. N. H. Adv. Synth. Catal. 2008, 350, 1610–1614.
(c) Wassenaar, J.; Reek, J. N. H. J. Org. Chem. 2009, 74, 8403–8406.
(15) (a) Wassenaar, J.; van Zutphen, S.; Mora, G.; Le Floch, P.; Siegler,
M.; Spek, A. L.; Reek, J. N. H. Organometallics 2009, 28, 2724–2734. (b)
Wassenaar, J.; Jansen, E.; van Zeist, W.-J. ; Bickelhaupt, F. M.; Siegler, M. A.;
Spek, A. L.; Reek, J. N. H. Nat. Chem. 2010, 2, 417–421.
(16) Wassenaar, J.; de Bruin, B.; Siegler, M. A.; Spek, A. L.; Reek,
J. N. H.; van der Vlugt, J. I. Facile oxidation of Rh(I) and Ir(I) complexes
derived from new tripodal tetradentate ligands (INOR-260). Abstracts of
Papers, 238th ACS National Meeting, Washington, DC, United States, August
16-20, 2009.
(10) Venanzi reported chemistry with an aryl-bridged analogue of PP3
ligand A: Dawson, J. W.; Venanzi, L. M. J. Am. Chem. Soc. 1968, 90, 7229–
7233.
(11) Field, L. D.; Guest, R. W.; Vuong, K. Q.; Dalgarno, S. J.; Jensen, P.
Inorg. Chem. 2009, 48, 2246–2253.
(12) (a) Foltz, C.; Enders, M.; Bellemin-Laponnaz, S.; Wadepohl, H.;
Gade, L. H. Chem.;Eur. J. 2007, 13, 5994–6008. (b) Ward, B. D.; Bellemin-
Laponnaz, S.; Gade, L. H. Angew. Chem., Int. Ed. 2005, 44, 1668–1671.
(c) Bellemin-Laponnaz, S.; Gade, L. H. Angew. Chem., Int. Ed. 2002, 41, 3473–
3475.
(13) We prefer the term flexidentate over hemilabile, as the former more
strongly implies bidirectional coordination character.

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6497
Figure 2. Recently developed hybrid ligand class IndolPhos,14 the corresponding phospholane-phosphoramidite used in asymmetric allylic alkylation,15
and Lahuerta’s tris(indolylphosphine)methane ligand.6.
Scheme 1. Synthesis of Tripodal Ligands 1-3a
a Conditions: (i) n-BuLi, THF, -78 °C; CO2; t-BuLi; ClPR2 (R = Ph, iPr). (ii) n-BuLi, THF, -78 °C; PCl3. (iii) n-BuLi, THF, -78 °C; CO2; t-BuLi;
PCl3. (iv) n-BuLi, THF, -78 °C; ClPPh2.
in a 2:1:1 ratio. One iPr methyl group is shifted to high field
species 1 and the tris-oxide of 3 were grown from CDCl3-
Et2O and THF-hexane, respectively (see Supporting In-
formation). For compound 1,16 the obtained intramolecular
(δ -0.06 ppm), likely a result of anisotropic shielding from
the aromatic indole rings. An independent synthesis to obtain
17
ꢀ
ligand 1 was recently reported by the group of Perez-Prieto,
distances and angles are very similar to the structure reported
17
ꢀ
albeit that the original, unoptimized procedure for the pre-
paration of 2-diphenylphosphino-3-methylindole was em-
ployed whereas the applied in situ CO2 protection metho-
dology reported for IndolPhos is superior.16
Conversely, ligand 3 bears a bridgehead phosphorus atom
that is connected at the 2-position of 3-methylindole, thus
reversing the overall pattern and synthetic methodology
(Scheme 1).18 Capping of PCl3 with 3 equiv of 2-lithioindole,19
employing in situ protection of the nitrogen using CO2,20
followed by phosphorylation of the nitrogen gave the desired
product3, the linkage isomer of 1, in good yield, with chemical
shifts at δ -75.1 (quartet) and 37.3 ppm (doublet), with a
coupling constant JP-P of 159 Hz.
The two linkage isomers show remarkable differences in
the NMR spectra; the Δδ for the pivotal-phosphorus atom is
∼150 ppm, while the peripheral phosphines (PPh2 only) span
a range of 66 ppm between ligand 1 and 3. This suggests
strikingly different geometric and/or electronic environments
for the two types of P-atoms in each ligand. Single crystals for
by Perez-Prieto. Oxidation of the peripheral phosphines
occurred in solution during the slow crystallization of 3.
Although this compound is stable when stored as a solid, this
observation might indicate a subtle difference in the electro-
nic properties of the phosphines in1 and 3. The C3-symmetric
nature of these ligand scaffolds is immediately apparent, as
well as the helical feature enforced by the sterically encum-
bered side groups. The lone pair on the central P1 atom points
inward with respect to the cavity created by the three bulky
arms. The overall C3-symmetry and related basket-shaped
configuration likely also dominates in solution as a result of
the rigid ligand architecture. This is indicated by the large
coupling constant JP-P between the inequivalent P-atoms,
with values of 197 and 159 Hz for 1 and 3, respectively.
Coordination Chemistry to CuI. We initially focused on
Cu as the metal of choice for these PP3 ligands, to
investigate its coordination chemistry as well as potential
catalytic applications (Scheme 2).
Reaction of ligands 1-3 with [Cu(cod)(μ-Cl)]2 led to
the light-yellow solids [CuCl(1)] (4), [CuCl(2)] (5), and
[CuCl(3)] (6) in moderate to excellent yields, which
showed some broadened signals in the NMR spectra.
The 31P NMR spectrum of 4 showed a quartet at δ 58.70
and a doublet at -27.68 ppm, with a JP-P coupling
constant of 198 Hz, suggesting on average C3-symmetry
in solution on the NMR time scale at rt. Cooling of the
sample led to broadening of all signals, but unfortunately
the dynamic coordination process could not be frozen
out. At -60 °C, complete coalescence of the signals was
still not observed in CD2Cl2. Similar spectral features at rt
(17) During the final writing stage of our current paper, a Note appeared
with the independent synthesis of compound 1 as well as some coordination
chemistry with Pd and characterization of the Rh(Cl)-complex (similar to
ꢀ
our species 7): Penno, D.; Koshevoy, I. O.; Estevan, F.; Sanau, M.; Angeles
ꢀ
Ubeda, M.; Perez-Prieto, J. Organometallics 2010, 29, 703–706.
(18) A preliminary account of this work has been communicated recently:
Wassenaar, J.; de Bruin, B.; Siegler, M. A.; Spek, A. L.; Reek, J. N. H.; van
der Vlugt, J. I. Chem. Commun. 2010, 46, 1432–1434.
(19) Tris-2-(3-methylindolyl)phosphine has been reported, see: Yu, J. O.;
Browning, C. S.; Farrar, D. H. Chem. Commun. 2008, 1020–1022.
(20) Katritzky, A. R.; Akutagawa, K.; Jones, R. A. Synth. Commun. 1988,
18, 1151–1158.

6498 Inorganic Chemistry, Vol. 49, No. 14, 2010
Wassenaar et al.
Figure 3. ORTEP plots (50% probability displacement ellipsoids) of CuI-complexes 4 (left), 5 (middle), and 6 (right). Hydrogen atoms and solvent
˚
molecules are omitted for clarity. Selected bond lengths (A) and angles (deg), for 4: Cu1-P1 2.2628(5), Cu1-P2 2.2468(6), Cu1 P3 3.7163(6), Cu1-P4
3 3 3
2.3292(5), Cu1-Cl1 2.2277(5); P1-Cu1-Cl1 116.93(2), P2-Cu1-Cl1 120.17(2), P4-Cu1-Cl1 123.82(2), P1-Cu1-P2 118.02(2), P1-Cu1-P4 83.81(2),
P2-Cu1-P4 83.91(2). For 5: Cu1-P1 2.2645(5), Cu1-P2 2.2632(5), Cu1 P3 3.8296(6), Cu1-P4 2.2994(5), Cu1-Cl1 2.2645(5); P1-Cu1-Cl1 105.78(2),
3 3 3
P2-Cu1-Cl1 115.36(2), P4-Cu1-Cl1 133.52(2), P1-Cu1-P2 129.12(2), P1-Cu1-P4 86.65(2), P2-Cu1-P4 85.99(2). For 6: Cu1-P1 2.2598(5), Cu1-P2
2.2689(5), Cu1 P3 3.9083(5), Cu1-P4 2.2841(7), Cu1-Cl1 2.2436(7); P1-Cu1-Cl1 115.82(3), P2-Cu1-Cl1 116.43(2), P4-Cu1-Cl1 125.02(2),
3 3 3
P1-Cu1-P2 120.79(2), P1-Cu1-P4 85.90(2), P2-Cu1-P4 86.16(2).
Scheme 2. Complexation of Novel Tripodal Ligands 1-3 to CuI
and -60 °C were observed for complex 5 and 6. Single
crystals suitable for X-ray crystallographic analysis were
obtained from CH2Cl2-toluene/hexane for complex 4-6
(see the molecular structures in Figure 3).
Complex 5 exhibits an even more distorted geometry
indicated by a large Pcentral-Cu-Cl angle of 133.52(2)°.
The slightly more flexible nature of the ligand of complex 6
allows for a more open geometry, which is indicated by a
The CuI ions in all complexes adopt distorted tetra-
hedral geometries, with coordination to two peripheral
PR2 groups, the central phosphorus atom of the tripodal
ligand and to the chloride coligand. The total sums of all
angles are 647° (4), 656° (5), and 650° (6), respectively.
Tetradentate coordination of the ligands is unfavorable
as this would lead to 20 electron complexes. Hence, there
is no element of symmetry left in the molecules, as one
phosphine unit is dangling in the solid state, while there is
rapid exchange of the peripheral donor units in solution
according to the NMR studies. The Cu-P distances in
˚
larger Cu P3 distance of 3.9083 A. This may be the
3 3 3
reason for the lower selectivity in the CuI catalyzed cyclo-
propanation as the Cu ion is less shielded by the ligand
framework (vide infra).
CuI Catalyzed Cyclopropanation. A first indication of
the catalytic applicability of the Cu-complexes 4-6 was
obtained by examining the Cu-catalyzed cyclopropana-
tion of styrene with ethyldiazoacetate (EDA) after in situ
removal of the chloride using AgOTf (Table 1).22 All
complexes gave active catalysts for this reaction and
quantitative conversion (based on EDA) after 18 h reac-
tion time, with the Cu-catalyst containing the more rigid
ligand 1 giving the highest trans to cis ratio of 70/30
(complex 4). However, significant amounts (up to 43%)
of diethyl maleate and fumarate are formed as a result of
Cu-catalyzed dimerization of EDA. Interestingly, com-
plex 6, containing the linkage isomeric ligand 3, gives
lower amounts of dimers (28%, entry 3). This shows
that the subtle differences in ligand structure between
these linkage isomers lead to altered reactivity of the
resulting metal complexes. Control experiments with
˚
complex 4 vary from 2.2468(5) A for Cu-Pperiphery to
2.3292(5) A for Cu-Pcentral, in the range of typical CuI-
˚
phosphine complexes.21 The remaining Cu P3 distance
3 3 3
˚
˚
is 3.7163(6) A and Cu-Cl is 2.2277(5) A. The two
Pcentral-Cu-Pperiphery angles are 83.81(2) and 83.91(2)°,
while the Pcentral-Cu-Cl angle is much larger at
123.82(2)°. This again is a clear indication of the con-
strained geometry of the formally tripodal ligand scaffold.
(21) For recent examples, see: (a) van der Vlugt, J. I.; Pidko, E. A.; Vogt,
D.; Lutz, M.; Spek, A. L.; Meetsma, A. Inorg. Chem. 2008, 47, 4442–4444.
(b) van der Vlugt, J. I.; Pidko, E. A.; Vogt, D.; Lutz, M.; Spek, A. L. Inorg. Chem.
2009, 48, 7513–7515. (c) Sircoglou, M.; Bontemps, S.; Mercy, M.; Miqueu, K.;
Ladeira, S.; Saffon, N.; Maron, L.; Bouhadir, G.; Bourissou, D. Inorg. Chem.
2010, 49, 3983–3990.
[CuOTf]2 benzene give a 60/40 trans/cis ratio and only
12% dimer formation (entry 4). Apparently the ligands
3
(22) (a) Davies, H. M. L.; Hedley, S. J. Chem. Soc. Rev. 2007, 36, 1109–
1119. (b) Kirmse, W. Angew. Chem., Int. Ed. 2003, 42, 1088–1093.

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6499
Table 1. Catalytic Cyclopropanation of Styrene and EDA (Ethyl Diazoacetate) Using New CuI-Complexes 4-6a
entry
catalyst
conversionb
yieldb,c
% trans/cis ratiob
% dimersb
1
2
3
4
[Cu(1)]OTf (4Ph
)
)
100
100
100
100
57
66
72
88
70/30
69/31
66/34
60/40
43
34
28
12
[Cu(2)]OTf (5iPr
[Cu(3)]OTf (6)
[CuOTf]2 benzene
3
1
a Ratio styrene:EDA:catalyst =800:100:1; temperature =25 °C; reaction time =18 h; solvent: CH2Cl2. b Determined by H NMR spectroscopy.
c Combined yield of cis and trans-isomers
Scheme 3. Coordination of Ligands 1 and 3 toward Rh and Ir Precursors, Leading to Complexes 7-12
enforce a slight increase in selectivity toward the trans
product, but further ligand design will be necessary to
enable a semi-enclosed coordination of the Cu-center
with potentially altered reactivity and selectivity.
Therefore, to further investigate the potential of the
new linkage isomeric ligands 1 and 3, each ligand was
reacted with suitable Rh and Ir precursors (Scheme 3).
Complexation with either neutral Rh or Ir was achieved
by reaction of stoichiometric amounts of the respective
ligand with [Rh(cod)2(μ-Cl)]2 or [Ir(cod)2(μ-Cl)]2. The
resulting complexes 7 [RhCl(1)],17 8 [RhCl(3)], 9 [IrCl(1)],
and 10 [IrCl(3)] were fully characterized by NMR spec-
troscopy, elemental analysis, and high resolution mass
spectrometry. It should be noted that formation of com-
pound 10 was sluggish, compared to the other three species.
The 31P NMR spectrum for 7 consists of a doublet-of-
quartets at δ 131.8 ppm (Δδ is 58.1 ppm) for the central
P-atom (JRh-P coupling constant of 170 Hz), and the
combined peripheral phosphorus-units are present at δ
5.7 ppm as a doublet-of-doublets, with a JP-P coupling
constant of 30 Hz (JRh-P: 138 Hz). Surprisingly, attempts
to coordinate the iPr analogue (ligand 2) to, for example,
the Rh-precursor failed, presumably because of excessive
steric crowding around the metal center.
Coordination Chemistry to Rh and Ir. Strikingly,
C3-symmetric scaffolds have hardly been used with Rh
to date,4d,6a,17,23 yet the combination of a rigid ligand
structure and an active protected metal site might yield
interesting reactivity. The use of rigid ligand scaffolds is
attractive and necessary to enforce a specific metal geo-
metry, both to restrict and direct the desired reactivity of
the metal center. In conjunction with tetradentate coor-
dination, conformational freedom is minimized, thus
paving the way for rational design and targeted reactivity.
If encapsulation of the metal center by the C3-symmetric
ligand appears effective with the archetypical MI
oxidation state (M=Rh and Ir), such ligand frameworks
might also open up opportunities to synthesize and
stabilize the open-shell derivatives (MII) and to study
their characteristics.
Single crystals for both Rh-complexes 7 and 8 were
obtained by slow diffusion of hexane into a tetrahydro-
furan (THF) or dichloromethane (DCM) solution (see
Supporting Information for details). Both structures dis-
play a distorted trigonal bipyramidal geometry. The
respective Rh-P distances including the axially coordi-
(23) Rh-complexes with the aliphatic tripodal tetradentate ligand
P(CH2CH2PPh2)3 have been reported, see ref 9 and: (a) Di Vaira, M.; Ehses,
M. P.; Peruzzini, M.; Stoppioni, P. Eur. J. Inorg. Chem. 2000, 2193–2198.
(b) Heinekey, D. M.; van Roon, M. J. Am. Chem. Soc. 1996, 118, 12134–12140.
(c) Bianchini, C.; Elsevier, C. J.; Ernsting, J. M.; Peruzzini, M.; Zanobini, F.
Inorg. Chem. 1995, 34, 84–92. (d) Taqui Khan, M. M.; Martell, A. E. Inorg.
Chem. 1974, 13, 2961–2966.
ꢀ
˚
˚
nated pivotal P1-donor of 2.1124(4) A (7) and 2.1592(7) A

6500 Inorganic Chemistry, Vol. 49, No. 14, 2010
Wassenaar et al.
(8) are slightly shorter than those found for the equato-
rially coordinated side arm phosphines (at ∼2.29-
˚
2.36 A). As with the structure for ligand 1, independent
synthesis and solid-state structure determination of 7 was
ꢀ
reported by Perez-Prieto and co-workers.17 Their data are
very similar to ours, with subtle but noticeable differences
most likely reflecting slight differences in crystal packing
combined with different collection temperatures (room
temperature vs 110 K in our case).
Contrary to pyrrolyl-based phosphorus ligands, which
tend to act as ‘pseudophosphites’ and as a result show
longer Rh-P bond lengths compared to traditional
phosphine ligands,24 the respective bridgehead P-atom
in complexes 7 and 8 appear not to participate in
π-backdonation, thus resembling more classical P(NR2)3
donors.25 The latter is further evidenced by the absence of
elongation of the P1-N bonds in the complexes com-
pared to the free ligand. It should be noted that the
deviation from ideal tbp geometry structure is somewhat
larger for 7 (τ-value: 0.80) than for complex 8 (τ-value:
0.93).26 This can be explained by the even more rigid
ligand framework of 1.
Figure 4. DFT optimized geometry for complex 11, [Rh(1)(nbd)]BF4,
withan 2,3-η2-monodentatecoordinationofthe norbornadienefragment.
Unprecedented Monodentate 2,3-η2-nbd Coordination
to Rh. A cationic analogue of the RhCl-complex 7 with
ligand 1 was also attainable, by using [Rh(nbd)2]BF4
(nbd = norbornadiene) as precursor. This led to the forma-
tion of an orange-red colored species [Rh(1)(nbd)]BF4 (11),
which also exhibited a clear 31P NMR spectrum (δ 132.4
for hydrogen coordination. This is due to effective shield-
ing of the metal by the rigid ligand framework. Notwith-
standing this observation, substitution chemistry pro-
ceeded smoothly as the formation of, for example, the
corresponding CO-complex, [Rh(1)(CO)]BF4 (12), was
facile upon exposure of complex 11 under 25 bar CO at rt
in CH2Cl2, yielding the corresponding orange-colored
monocarbonyl complex. The 31P NMR spectrum indi-
cated a downfield shift of the peripheral phosphines to
16.3 ppm (Δδ = 10 ppm) compared to nbd-complex 11,
most probably resulting from the stronger π-acidic char-
acter of the CO ligand. The FT-IR spectrum showed one
strong band at νCO 2000 cm-1. This is in the expected
range for five-coordinated cationic Rh(CO)-species27 and
is somewhat higher than in the related [Rh(A)(CO)]BF4
complex (νCO 1985 cm-1), because of the electron-accept-
ing nature of the indole based ligand 1 compared to the
alkyl based ligand A.9 However, judging from the νCO
value, the indolylphosphine fragment is definitely not a
strong π-acceptor, unlike the pyrrolylphosphine ana-
logues. Single crystals of 12 suitable for X-ray crystal-
lography were obtained from slow diffusion of hexane
into a CH2Cl2 solution (Figure 5). The structure exhibits a
distorted trigonal bipyramidal geometry (τ-value: 0.75),
similar to the corresponding chloride complex 7. The
Rh1-P4 distance is 2.1978(6) A, which is 0.085 A longer
than in the more electron rich chloride complex 7. This
indicates that even though the ligand may be a weaker
π-acid compared to pyrrolyl-analogues, the pivotal phos-
phorus donor P4 does participate in π-backbonding.
Reactivity of RhI Complexes toward Acids and Alkylat-
ing Agents. In marked contrast with the aforementioned
work on the RhI complex of ligand A,9 we were unsuc-
cessful in obtaining a stable RhIII dihydrido-species or
subsequent monohydrido complexes with ligands 1 or 3,
by treatment of the cationic [RhIII(H)(Cl)]OTf species 13,
obtained by reaction of 7 with triflic acid, with excess
1
and 6.4 ppm) and a highly ordered H NMR spectrum,
with unambiguous signals for a single coordinated nbd
coligand with a rare 2,3-η2-monodentate coordination
mode. This is evidenced by the observation of two sets of
signals for the diene protons in the 1H NMR spectrum at
5.37 and 4.55 ppm for the η2 coordinated double bond, and
at 6.33 and 6.25 ppm for the non-coordinated double bond.
To our knowledge this is the first Rh(nbd)-complex where-
in the norbornadiene acts as a monodentate ligand, leaving
one double bond uncoordinated, which must be a result of
the coordinating power of the multidentate phosphorus
ligand. In addition, all diene signals are doubled, because of
the chiral C3-symmetric environment imposed by the tetra-
phosphine ligand. To further understand the bonding
nature in this species, we performed density functional
theory (DFT) calculations (see Figure 4). The optimized
structure shows just one coordinated double bond of the
nbd ligand, which is coordinated in the anticipated endo-
conformation. This is in agreement with NMR spectro-
scopic observations wherein only two diene protons are
shifted upfield. The intramolecular distance Rh-(C1dC2) for
˚
˚
˚
the coordinated double bond is approximately 2.3 A,
whereas the distance of the free double bond (C3dC4) to
˚
the Rh-center is beyond bonding range at ∼3.6 A in the DFT
optimized structure.
Substitution to Give Monocarbonyl Complex 12. Com-
plex 11 proved to be unreactive toward molecular hydro-
gen, as no hydrogenation of the diene was observed. This
can be understood by the lack of vacant sites on the metal
(24) Moloy, K. G.; Petersen, J. L. J. Am. Chem. Soc. 1995, 117, 7696–
7710.
(25) Crabtree, R. H. In The Organometallic Chemistry of the Transition
Metals, 3rd ed.; John Wiley & Sons Inc.: New York, 2001; pp 91-95.
(26) Addison, A. W.; Rao, T. N.; Reedijk, J.; van Rijn, J.; Verschoor,
G. C. J. Chem. Soc., Dalton. Trans. 1984, 1349–1356.
(27) (a) de Bruin, B.; Donners, J. J. J. M.; de Gelder, R.; Smits, J. M. M.;
Gal, A. W. Eur. J. Inorg. Chem. 1998, 401–406. (b) Gambaro, J. J.; Hohman,
W. H.; Meek, D. W. Inorg. Chem. 1989, 28, 4154–4159.

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6501
Scheme 4. Oxidation of RhI Complexes 7 and 8 to Form Stable
Cationic Metalloradical RhII Analogues
Spurred by initial attempts to abstract the chloride
ligand in 8 with Ag-salts, which resulted in an instant-
aneous color change from burgundy-red to blackish
concomitant with formation of a Ag0-mirror rather than
precipitation of anticipated AgCl, indicating redox-based
reactivity, we observed smooth oxidation of both complex
7 and 818 with ferrocenium hexafluorophosphate to yield
the one-electron oxidized PP3)RhII-Cl complexes [7]PF6
and [8]PF6 (Scheme 4). Although not unprecedented,32 the
isolation of stable RhII-chloro complexes is unexpected
becauseinothercasesthe presenceofCl- has been reported
to trigger disproportionation of otherwise stable RhII
species.33 In this case, disproportionation of (PP3)RhII-Cl
is likely hindered by the strained (ligand) geometries of the
hypothetical (trigonal pyramidal) (PP3)RhI and (cons-
trained octahedral) (PP3)RhIII-Cl products. Iridium com-
plexes 9 and 10 could also be oxidized by ferrocenium hexa-
fluorophosphate, but no stable IrII species could be isolated.
Electronic Absorption Spectraand TD-DFT Calculations.
The absorption spectra of 8 and [8]PF6 exhibit distinguish-
ably different features (Figure 6). The parent compound
exhibits bands at λ 475 nm (ε 3.6 ꢀ 103 M-1 cm-1) and λ
528 nm (ε 2.7 ꢀ 103 M-1 cm-1) in CH2Cl2. For [8]PF6 only
one absorption band at λ 550 nm (ε 2.5 ꢀ 103 M-1 cm-1).
Similar absorption spectra were obtained for complex 7
and its cation [7][PF6] (see Supporting Information).
In an attempt to understand the nature of the electronic
transitions we performed TD-DFT calculations with the
Orca program at the ri-DFT BP86 level (COSMO CH2Cl2),
using the Turbomole optimized geometries of 8 and 8þ.
These correctly predict the experimentally observed red
shifts and decreased intensities of the bands in the visible
range of the spectra after oxidation of 8 to 8þ. The energies
of the calculated transitions do not match the experimental
data very accurately, with most of the calculated transitions
occurring at roughly 100 nm (8) to 160 nm (8þ) wave-
numbers too high. This could perhaps be improved some-
what using a hybrid HF-DFT functional and a larger
basis set, but the current data are sufficient for a qualitative
understanding of the observed electronic transitions.
Figure 5. ORTEP plot (50% probability displacement ellipsoids) of the
cationic portion of complex 12. Hydrogen atoms, the BF4 anion, and
˚
solvent molecules are omitted for clarity. Selected bond lengths (A) and
angles (deg): Rh1-P1 2.3545(6), Rh1-P2 2.3174(7), Rh1-P3 2.1978(6),
Rh1-P4 2.1978(6), Rh1-C64 1.925(3), C64-O1 1.119(3); P1-Rh1-C64
108.84(7), P2-Rh1-C64 92.35(7), P3-Rh1-C64 95.64(7), P4-Rh1-C64
168.64(7), P1-Rh1-P4 82.52(2), P2-Rh1-P4 81.75(2), P3-Rh1-P4
79.81(2), Rh1-C64-O1 172.6(2).
NaBH4 (see Supporting Information). Furthermore, RhCl-
complexes 7 and 8 have failed to show any appreciable
reactivity toward methyl- or phenyllithium, because of
an unusual stability of the Rh-Cl bond, hence preclud-
ing the synthesis of species featuring a Rh-carbon
bond. This is in contrast with the results reported with
ligand A, where Cl could be displaced for alkyl frag-
ments via salt metathesis. This could be an indication
that the trans influence of the phosphorus bridgehead in
1 and 3 is small compared to the central alkylphosphine
fragment in A.
Oxidation to RhII. Rh-chemistry is dominated by the
coordination chemistry of RhI and RhIII, while the inter-
mediate oxidation state of RhII is largely considered
exotic.28 However, a growing body of evidence is suggesting
that such odd-electron species might play an important role
in the transformation of organic compounds.29 In fact,
dinuclear bis-RhII species catalyze the (Doyle-type) cyclo-
propanation of alkenes with diazo compounds.30 Mono-
nuclear paramagnetic RhII and IrII species are studied to a
lesser extent, but these compounds also reveal interesting
reactivity in the single-electron activation of small mole-
cules such as CO or diazo compounds.31 Isolation and
characterization of these MII complexes is challenging
because of their inherent radical reactivity.
(28) For reviews on mononuclear RhII- and IrII-species, see: (a) DeWit,
D. G. Coord. Chem. Rev. 1996, 147, 209–246. (b) Hetterscheid, D. G. H.;
Gr€utzmacher, H.; Koekoek, A. J. J.; de Bruin, B. Prog. Inorg. Chem. 2007, 55,
247–354. (c) de Bruin, B.; Hetterscheid, D. G. H. Eur. J. Inorg. Chem. 2007,
211–230.
(29) de Bruin, B.; Budzelaar, P. H. M.; Gal, A. W. Angew. Chem., Int. Ed.
2004, 43, 4142–4157.
(30) (a) Davies, H. R. L.; Manning, J. R. Nature 2008, 451, 417–424.
(b) Davies, H. R. L.; Hedley, S. J. Chem. Soc. Rev. 2007, 36, 1109–1119.
(c) Merlic, C. A.; Zechman, A. L. Synthesis 2003, 1137–1156.
(31) (a) Dzik, W. I.; Smits, J. M. M.; Reek, J. N. H.; de Bruin, B.
Organometallics 2009, 28, 1631–1643. (b) Dzik, W. I.; Reek, J. N. H.; de Bruin,
B. Chem.;Eur. J. 2008, 14, 7594–7599.
(32) Krumper, J. R.; Gerisch, M.; Suh, J. M.; Bergman, R. G.; Tilley,
T. D. J. Org. Chem. 2003, 68, 9705.
(33) (a) Hetterscheid, D. G. H.; Smits, J. M. M.; de Bruin, B. Organome-
tallics 2004, 23, 4236–4246. (b) Sun, H.; Xue, F.; Nelson, A. P.; Redepenning, J.;
DiMagno, S. G. Inorg. Chem. 2003, 42, 4507.

6502 Inorganic Chemistry, Vol. 49, No. 14, 2010
Wassenaar et al.
Figure 6. UV-vis spectra of complex 8 and [8]PF6 in CH2Cl2 (left), together with TD-DFT simulated spectra (right).
The lowest energy bands of RhI compound 8(Exp: 528 nm,
475 nm; DFT: 625 nm, 588 nm) are transitions between
(delocalized) MOs build mainly from σ-interactions between
different metal d-orbitals and the lone pairs at the phosphorus
ligand donors (HOMO f LUMO and HOMO-1 f
LUMO), so these are basically “d-d transitions” with the
P-lone pairs strongly (covalently) mixed in. The higher energy
transitions (e.g., 454 nm in DFT; UV range in the experi-
mental spectrum) are mainly π f π* transitions. The lowest
energy transitions of RhII compound 8þ (Exp: 550 nm;
DFT 710 nm, 716 nm) are transitions occurring between
(delocalized) MOs built mainly from σ-interactions between
different metal d-orbitals and the lone pairs at the phosphorus
and chloro ligand (SOMOfR-LUMO and β-HOMO-5 f
β-LUMO = SOMO), so again essentially “d-d transitions”,
but with the phosphorus and chloride lone pairs strongly
(covalently) mixed in. The higher energy transitions (i.e., the
Figure 7. Cyclic voltammograms for complexes 7 (black line), 8 (red
line), 9 (green lines), and 10 (blue line) in CH2Cl2 vs Fc/Fcþ using 0.1 mM
nBu4PF6, showing only the MI/MII oxidation/reduction potential. Scan
transitions between 650 and 450 nm in the DFT calculated
spectrum) are dominated by mixed LMCT, MLCT, and
π-π* charge transfer bands (most of which are strongly
mixed orbitals, involving linear combinations of several vir-
tual orbitals of 8þ).
rate v = 0.1 V s-1
.
Table 2. Cyclic Voltammetric Data for Complexes 7-10 and Reference
Compound [RhCl(A)]a
Electrochemistry. Cyclic voltammetry for complexes
7-10 was carried out to investigate the redox-chemistry
of the electron-rich RhI and IrI complexes (Figure 7). As a
reference complex the known species Rh(P(CH2CH2-
PPh2)3)Cl9 [RhCl(A)], based on alkylphosphine ligand
A, was chosen for comparison (Table 2). All complexes
show (almost) fully reversible MI/MII oxidation waves at
potentials ranging from -0.9 ([RhCl(A)]) to -0.4 V (8),
referenced vs Fc/Fcþ. The relatively low potentials ob-
served indicate facile oxidation, which likely results from
the tetradentate coordination of four strongly electron-
releasing phosphines, making the metal center electron-
rich. Indeed, the oxidation potential is lowest for refer-
ence complex [RhCl(A)],9 containing one alkyl- and three
complex
E1/2 (V )
ΔE (V )
I/Io
7
8
9
-0.58
-0.42
-0.61
-0.51
-0.86
0.13
0.13
0.11
0.12
0.11
1.00
0.85
0.95
0.97
0.97
10
[RhCl(A)]9
a Referenced vs Fc/Fcþ using 0.1 mM nBu4PF6, scan rate v = 0.1 V s-1
.
arylphosphine units, and highest for 8, containing one
arylphosphine and three phosphine amides.
The iridium complexes 9 and 10 show slightly lower
oxidation potentials compared to their Rh-analogues (for
9: Δ = 0.029 V; for 10: Δ = 0.090 V), which is in accord

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6503
Figure 8. Experimental (black) and simulated (red) X-band EPR spectra of [7]PF6 (left) and [8]PF6 (right)18 in frozenacetone. TBAH (∼0.1 M) was added
to obtain a better glass. Experimental details for [7]PF6: 40K, frequency 9.3807 GHz, microwave power 0.2 mW, modulation amplitude 2 G. Experimental
details for [8]PF6: 20K, frequency 9.3794 GHz, microwave power 0.2 mW, modulation amplitude 4 G. The simulation parameters are listed in Table 3.
with previous observations.28 As expected, solution-state
RT magnetization measurements by Evans’ method34 are
consistent with the formulation of complexes [7]PF6 and
[8]PF6 as radical species, with μeff values of 2.06(3) μB
([7]PF6) and 2.29(1) μB ([8]PF6). These values are some-
what larger than the spin-only values (1.73 μB), thus
indicating spin-orbit influences, and they are in the same
range as PNP-based mononuclear RhII species reported
by Milstein and co-workers.35 The solution phase Evans’s
method is not very accurate, and the obtained values are
somewhat overestimated. RT susceptibility values ob-
tained in the solid state (Evans balance) for related
NNN-based RhII complexes36 are somewhat smaller (in
the range 1.6-1.95 μB), and the X-band electron para-
magnetic resonance (EPR) spectra of [7]PF6 and [8]PF6
indeed reveal small orbital contributions to the magnetic
susceptibility.
EPR Spectroscopic Analysis and DFT Calculations of
Metalloradical Species. The X-band EPR spectra of
[7]PF6 and [8]PF618 are shown in Figure 8. Identical data
were obtained for stored (one-month old) and in situ
generated samples, indicative of highly stable radical spe-
cies. The EPR parameters of the DFT optimized geo-
metries of both 7þ and 8þ were also calculated (BP86, TZP).
Simulation of the experimental EPR spectra of [7]PF6
(20 K) and [8]PF6 (40 K) reveal rhombic g-tensors with
moderate g-anisotropy and resolved superhyperfine coup-
lings with mainly three P atoms; large coupling with P2 and
smaller couplings with P3 and P4 (Table 2). The fourth
P-atom (pivotal P1) and Rh show no resolved hyperfine
couplings.
presence of many close-lying exited states, as has been
observed for IrII(por) species.37
The experimentally derived EPR parameters are in rea-
sonable agreement with those calculated with DFT (see
Table 3), and thus the geometries of 7þ and 8þ in frozen
solution should be very close to the DFT optimized struc-
tures, which in turn should resemble those present in the
solid state, as was established for the X-ray structure of 8þ
(see Supporting Information). A distorted geometry in-
between a trigonal bipyramid (tbp) and a square pyramid
(sqpy) is revealed. This is quite remarkable, because
five-coordinate RhII-complexes generally adopt a square-
pyramidal geometry (Jahn-Teller effect).34b
The geometry of 8þ, both in the solid and in (frozen)
solution, is thus best described as distorted square pyrami-
dal (sqpy), with the pivotal P donor (P1), the chloro ligand,
and two of the side arm P donors (P3 and P4) in the square
plane and the third side arm P donor at the apical position
(P2). However, this geometry is strongly distorted from the
ideal sqpy geometry toward a tbp, because the ligand is
preshaped to form the latter and therefore does not accom-
modate the appropriate angles for a sqpy. The complex
thus adopts a distorted geometry in-between the geome-
tries preferred by either ligand and metal. On the basis of
the similarity of the EPR spectra (Figure 8, Table 3), 7þ
should have a similar coordination geometry.
Because of its distorted geometry, the DFT spin-carry-
ing orbital (singly occupied molecular orbital, SOMO) of
8þ is not the expected metal dz2 orbital normally encoun-
0
tered in sqpy d7 complexes (dxzþdyz4, dxy2, dz21, dx2-y2
configuration), but a strongly delocalized orbital con-
structed of a different Rh d-orbital (possibly the “dxy” or
“dx2-y2”), in bonding combination with lone pairs of two
Interestingly, the corresponding in situ oxidized Ir-
complexes did not reveal any EPR-signals. This could
mean that the IrII species are inherently unstable (only
detectable on the fast CV time scale), but its EPR silence
could also be due to very fast relaxation caused by the
side arm P donors. The rather small g-anisotropy
4
excludes a (weakly distorted) tbp complex (dxzþdyz
,
dxyþdx2-y23, dz21, dz2 configuration), because (close to)
degenerate dxy and dx2-y2 orbitals should lead to a much
larger g-anisotropy (gzz . ge).28b The distorted geometry
actually leads to strong mixing of the d-orbitals among
several of the lower and higher lying orbitals, which pre-
vents a proper assignment of the “d-orbital configuration”
0
(34) Evans, D. F. J. Chem. Soc. 1959, 2003–2005.
(35) Feller, M.; Ben-Ari, E.; Gupta, T.; Shimon, L. J. W.; Leitus, G.;
Diskin-Posner, Y.; Weiner, L.; Milstein, D. Inorg. Chem. 2007, 46, 10479–
10490.
(36) See ref 28 and: Hetterscheid, D. G. H.; Klop, M.; Kicken, R. J. N. A.
M.; Smits, J. M. M.; Reijerse, E. J.; de Bruin, B. Chem.;Eur. J. 2007, 13,
3386–3405.
(37) Zhai, H.; Bunn, A.; Wayland, B. B. Chem. Commun. 2001, 1294–
1295.

6504 Inorganic Chemistry, Vol. 49, No. 14, 2010
Wassenaar et al.
Table 3. Experimental and DFT Calculated g-Values, Hyperfine Coupling
Constants, A [MHz], and Spin Populations F of [7]PF6 and [8]PF6
of 8þ on the basis of the DFT calculations. For both 7þ and
8þ, about 50% of the DFT spin density resides on Rh and
about 25% on the “apical” P2-donor (Figure 9, Table 3).
The species are thus best described as strongly covalent
radicals. They have substantial metallo-radical character,
but with a strong contribution of the sqpy “axial” P2 donor.
Reactivity of Metalloradicals toward Activation of H2.
We previously reported on the remarkable capability of
[8]PF6 to activate H2.18 When this complex was exposed
to molecular hydrogen for 24 h (5 bar, CD2Cl2), RhIIIH
species [Rh(3)(H)Cl]PF6 was formed quantitatively in
what appeared to be a direct oxidation-reaction to split
H2 into two protons and two electrons with subsequent
recombination. We thus wondered if the linkage isomeric
compound [7]PF6, showing subtly different redox-beha-
vior, would demonstrate similar reactivity (Scheme 5).
Indeed, dihydrogen activation takes place but at lower
rates than with [8]PF6 under identical conditions. Full
conversion toward [Rh(1)(H)Cl]PF6 (13) is achieved after
6 days instead of 24 h. Complex [7]PF6 shows a lower
reduction potential (at -0.58 V for the RhI/RhII couple)
compared to [8]PF6 (-0.42 V) (see Figure 7 and Table 2).
Hence, the redox-potential of [7]PF6 vs NHE is only 0.12
V vs 0.28 V for [8]PF6.18 These values imply that the
overall oxidation of H2 to 2Hþ, which is thermodynami-
cally allowed (even without subsequent metal proto-
nation), should indeed be possible. Furthermore, it
provides a rationale for the observed slower reaction with
complex [7]PF6, as it shows a lower overpotential. We
postulate an outer-sphere redox-reaction involving full
oxidation of H2 and consecutive protonation of the
generated RhI-complex. The difference in rate supports
our proposed mechanism in which dihydrogen is fully
reduced by the metalloradical complex over alternative
mechanisms such as discussed by Rauchfuss and co-
workers, involving oxidative deprotonation of an Ir(H2)
fragment with a non-innocent ligand radical,38 or via
homolytic H2 bond cleavage in a trimolecular process,
analogous to reported RhII(por) or Co(CN)53- systems.39
Given the very sterically encumbered nature of the ligand
and the resulting pseudoencapsulation of the metal cen-
ter, the latter mechanism seems very unlikely.
a
In the 31P NMR spectrum of 13 an AM2X spin pattern
could be discerned while the 1H NMR spectrum showed a
complex doublet-of-doublet-of-doublet-of-triplets at δ
2
-6.49 ppm, with a large JP-H coupling of 167.6 Hz.
The hydrido ligand is trans to one of the side arm P-atoms
and the chloro ligand trans to the pivotal P donor.
Independent synthesis of analogous salts of 13 confirmed
the above assignment.40
Initial attempts to further utilize the radical character
of the RhII complexes 7þ or 8þ have so far not been
(38) Another pathway for H2 activation was recently reported, see:
Ringenberg, M. R.; Latha Kokatam, S.; Heiden, Z. M.; Rauchfuss, T. B.
J. Am. Chem. Soc. 2008, 130, 788–789.
(39) H2 activation by RhII-porphyrins and RhII-salen complexes is well-
documented, see: (a) Cui, W.; Wayland, B. B. J. Am. Chem. Soc. 2004, 126,
8266–8274. (b) Zhang, X. X.; Wayland, B. B. J. Am. Chem. Soc. 1994, 116,
7897–7898. (c) Bunn, A. G.; Wei, M.; Wayland, B. B. Organometallics 1994, 13,
3390–3392. (d) Wayland, B. B.; Ba, S.; Sherry, A. E. Inorg. Chem. 1992, 31, 148–
150. (e) James, B. R.; Stynes, D. V. J. Am. Chem. Soc. 1972, 94, 6225–6226.
(40) Similar RhIII species have been made via classic routes, e.g., from
complex [Rh(A)Cl], see ref 9. For the related tris(indolylphosphino)methane
ligand reported by Lahuerta, see ref 6a.
a DFT values are given in italics. NR = not resolved. The experi-
mental values are least-square “best fit” values obtained by simulation of
the experimental X-band EPR spectra.

Article
Inorganic Chemistry, Vol. 49, No. 14, 2010 6505
Figure 9. SOMO (left) and spin density (right) of complex 8þ (view along the Cl-Rh- P1 axis). Plots for 7þ are similar.
Scheme 5. Formation of Cationic RhIII(H)Cl-Species 13 via Oxida-
tion of H2 by [7]PF6, and Subsequent Protonation of the RhI Species
on the overall structure and its reactivity. The RhII-com-
plexes were characterized by EPR-spectroscopy and for
complex [8]PF6 also by X-ray crystallography. The pro-
nounced overall C3-symmetry dictated by the very rigid
ligand geometry was also preserved in this corresponding
oxidation product.
The isolated metalloradical complexes 7þ and 8þ are rare
RhIICl compounds with an unusual coordination geometry in
between a trigonal bipyramid and a square pyramid. These
species appear bench-stable as solids, and they also do not
show appreciable decomposition in common solvents, indi-
cating highly effective shielding of the RhII center by the
ligand sphere. Notwithstanding this, both 7þ and 8þ react
with molecular hydrogen to quantitatively form the corre-
sponding cationic RhIII hydridochloride complexes. The
slower reaction with 7þ, which has a lower oxidation potential
compared to 8þ (Δ = 0.16 V), further supports the proposed
mechanism via complete outer-sphere oxidation of molecular
hydrogen to protons. Further studies aimed at the reactivity
of these unusual RhII metallo-radicals are ongoing, including
studies derived from the related cationic RhI species.
This detailed study of the coordination properties of the
novel PP3 frameworks 1-3 illustrates that indeed unusual
geometries of metal complexes as well as novel reactivity
of these complexes can be achieved. This is the result of the
highly rigid nature of the ligands in conjunction with the
encapsulating effect of the tripodal coordination mode, which
concurrently stabilize reactive species and direct their reactiv-
ity. It is remarkable to observe that these rigid PP3 ligands can
accommodate tetradentate complexation to Rh in three
oxidation states (Figure 10), with subtle but distinct differ-
ences in the overall structure. This concept will be further
expanded on for further reactivity studies, including with
other metals as well. Also the activation of small molecules
other than H2 will be targeted, including N2 and NH3.42
successful. We have investigated possibilities to apply the
observed activation of dihydrogen in hydrogen atom-
transfer reactions, using the RhII metalloradicals based
on these novel and modular tripodal phosphorus ligands
as mediators or initiators. No reaction was observed with
the H-atom transfer41 (HAT) donor 9,10-dihydroanthra-
cene or with TEMPO as a HAT-acceptor reagent. We will
intensify these efforts, coupled with strategies to modu-
late the periphery of our tripodal ligand system to en-
hance reactivity.
Conclusions
Summarizing, we have reported the straightforward synth-
esis of three novel tripodal, potentially tetradentate phos-
phorus ligands 1-3, derived from 3-methylindole as building
block. The corresponding CuICl-complexes 4-6 display
coordination of only three phosphorus atoms and give a
7:3 trans/cis selectivity in the cyclopropanation of styrene
with EDA. However, the complementary, linkage isomeric
PP3 frameworks enforce C3-symmetry onto neutral MI com-
plexes 7-10 (M = Rh, Ir), as deduced from NMR spectro-
scopic and X-ray crystallographic data. Cationic RhI
complexes were obtained with nbd (11) and CO (12) as a
co-ligand. Norbornadiene complex 11 displays monodentate
coordination of the diene ligand, which is hitherto unprece-
dented, to the best of our knowledge. Cyclic voltammetry on
the chloride complexes 7-10 showed facile oxidation to the
corresponding MII species and a strong dependence of the
oxidation potential on the overall ligand structure. Reference
compound [RhCl(A)], which features a more flexible and
electron-rich aliphatic PP3 scaffold, proved significantly
easier to oxidize. Hence, the inherent, designed rigidity
built-in for ligand 1 and 3 has a strong governing influence
Experimental Section
General Procedures. All reactions were carried out under an
atmosphere of argon using standard Schlenk techniques. With
exception of the compounds given below, all reagents were
purchased from commercial suppliers and used without further
purification. Diphenyl(3-methyl-2-indolyl)phosphine,14 tris-
2-(3-methylindolyl)phosphine43 and RhCl(P(CH2CH2PPh2)3)9a
(41) (a) Wu, A.; Masland, J.; Swartz, R. D.; Kaminsky, W.; Mayer, J. M.
Inorg. Chem. 2007, 46, 11190, and references therein. (b) B€uttner, T.; Geier, J.;
(42) van der Vlugt, J. I. Chem. Soc. Rev. 2010, 39, 2302–2322.
(43) This compound has been synthesized before using an aminal protect-
ing route, see ref 15.
€
Frison, G.; Harmer, J.; Calle, C.; Schweiger, A.; Schonberg, H.; Gr€utzmacher, H.
Science 2005, 307, 235–238.

6506 Inorganic Chemistry, Vol. 49, No. 14, 2010
Wassenaar et al.
Figure 10. Space-filling models for complexes7 (left), [7]PF6 (middle), and 13(right), showing the verysterically encumbered butsubtlychanging situation
around the respective RhI, RhII, and RhIII-center, respectively. Complex 7 has been crystallographically characterized (see Supporting Information),
whereas [7]PF6 and 13 have been calculated by DFT.
were synthesized according to published procedures. THF,
pentane, hexane and diethyl ether were distilled from sodium
benzophenone ketyl; CH2Cl2, isopropanol, and methanol were
distilled from CaH2, and toluene was distilled from sodium
under nitrogen. Melting points were recorded on a Gallenkamp
melting point apparatus and are uncorrected. NMR spectra (1H,
31P and 13C) were measured on a Varian INOVA 500 MHz or a
Varian MERCURY 300 MHz. High resolution mass spectra
were recorded on a JEOL JMS SX/SX102A four sector mass
spectrometer; for FAB-MS 3-nitrobenzyl alcohol was used
as matrix. Elemental analyses were carried out at Kolbe
Tris-2-(3-methyl-N-diphenylphosphinoindolyl)phosphine (3).
To a solution of tris-2-(3-methylindolyl)phosphine (1.5 g, 3.56
mmol) in THF (50 mL) was added n-BuLi (2.5 M in hexanes,
4.28 mL, 10.7 mmol) at -78 °C. The resulting solution was
stirred for 1 h, and chlorodiphenylphosphine (1.92 mL, 10.7
mmol) was added. The reaction mixture was stirred for 16 h
allowing to warm slowly to rt. The resulting white suspension
was concentrated in vacuo and redissolved in CHCl3 (40 mL).
The suspension was filtered through a pad of basic alumina,
which was rinsed with CHCl3 (2 ꢀ 20 mL). The solvent was
removed under reduced pressure, and the resulting white solid
was washed with Et2O (3 ꢀ 10 mL) and dried in vacuo. Yield:
2.00 g (58%). Mp = 267 °C (dec.). 1H NMR (500 MHz; CDCl3,
298 K): δ 7.43 (d, J = 7.9 Hz, 3H), 7.24-7.10 (m, 30H), 7.02 (t,
J = 7.4 Hz, 3H), 6.83 (t, J = 7.7 Hz, 3H), 6.71 (d, J = 8.3 Hz,
3H), 2.05 (s, 9H). 13C NMR (126 MHz; CDCl3, 298 K): δ 140.6
(d, JCP = 10.7 Hz), 136.4-136.2 (m), 136.1 (d, JCP = 16.8 Hz),
€
Mikroanalytisches Laboratorium in Mulheim an der Ruhr.
Cyclic voltammograms were recorded on an Eco Chemie Auto-
lab PGSTAT10 potentiostat. UV-vis spectra were measured on
a Hewlett-Packard 8453 spectrophotometer. X-band EPR spec-
troscopy measurements were performed with a Bruker EMX
Plus spectrometer.
Tris-N-(3-methyl-2-diphenylphosphinoindolyl)phosphine (1).17 To
a solution of diphenyl(3-methyl-2-indolyl)phosphine (4.0 g, 12.7
mmol) in THF (60 mL) was added n-BuLi (2.5 M in hexanes, 5.08
mL, 12.7 mmol) at -78 °C. The resulting solution was stirred for
30 min and phosphorus trichloride (369 μL, 4.2 mmol) was added.
The reaction mixture was stirred for 2 days allowing to warm to rt.
The resulting white suspension was concentrated in vacuo and
redissolved in CHCl3 (50 mL). The suspension was filtered through
a pad of SiO2, which was rinsed with CHCl3 (2 ꢀ 50 mL). The
solvent was removed under reduced pressure and the resulting white
solid was washed with Et2O (3 ꢀ 20 mL) and dried in vacuo. Yield:
3.21 g (78%). Mp = 293 °C(dec.).1H NMR (500 MHz; CDCl3;298
K): δ 7.54 (d, J = 7.8 Hz, 3H), 7.28-7.26 (m, 3H), 7.19-7.14 (m,
9H), 7.11 (t, J = 7.6 Hz, 6H), 7.07-7.04 (m, 3H), 7.02-6.96 (m,
9H), 6.66 (d, J = 8.5 Hz, 3H), 6.51 (t, J = 7.2 Hz, 6H), 1.77 (s, 9H).
13C NMR (126 MHz; CDCl3; 298 K): δ 134.2 (d, JCP = 20.6 Hz),
133.8 (d, JCP = 30.9 Hz), 133.6-133.2 (m), 131.4 (d, JCP = 18.1
Hz), 128.4 (d, JCP = 13.2 Hz), 128.3-127.9 (m), 127.2, 125.8 (m),
124.9, 120.8, 119.4, 114.0, 10.7 (CH3). 31P{1H}-NMR (202 MHz;
CDCl3; 298 K): δ 73.72 (q, J = 197.3 Hz, 1P), -29.58 (d, J = 197.4
Hz, 3P). Anal. Calcd for C63H51N3P4 (%): C 77.69, H 5.28, N 4.31;
found: C 77.41, H 5.47, N 4.22. HRMS (FAB) calcd for [M þ H]þ
C63H52N3P4, 974.3112; found, 974.3117.
133.8-133.7 (m), 131.9 (t, JCP = 21.1 Hz), 128.8, 128.3 (t, JCP
=
5.5 Hz), 124.3-124.2 (m), 122.4, 120.0, 118.9, 114.5, 9.8 (CH3).
31P{1H}-NMR (202 MHz; CDCl3, 298 K): δ 37.27 (d, J = 159.2
Hz, 3P), -75.07 (q, J = 159.3 Hz, 1P). Anal. Calcd for the
trisphosphine oxide C63H51N3O3P4 (%): C 74.04, H 5.03, N
4.11; found: C 74.20, H 5.17, N 4.12. HRMS (FAB) calcd for
[M þ H]þ C63H52N3P4, 974.3112; found, 974.3108.
[Cu(1)Cl] (4). Tetraphosphine 1 (150 mg, 0.15 mmol) and
[CuCl(cod)]2 (32 mg, 0.08 mmol) are stirred in CH2Cl2 (8 mL)
for 1 h at rt. The resulting pale yellow solution was concentrated
under reduced pressure, and hexanes (10 mL) were added to
precipitate a pale yellow solid. The solid was collected by
filtration and washed with hexanes (5 mL). Crystals suitable
for X-ray single crystal diffraction were obtained by slow
diffusion of toluene in a CH2Cl2 solution at rt. Yield: 128 mg
1
(77%). H NMR (500 MHz; CD2Cl2; 298 K): δ 7.33 (m, 9H),
7.30 (d, J = 7.2 Hz, 9H), 7.22 (s, 6H), 7.10 (t, J = 7.3 Hz, 3H), 6.97
(q, J = 7.7 Hz, 9H), 6.70 (t, J = 7.8 Hz, 3H), 6.28 (d, J = 8.5 Hz,
3H), 1.56 (s, 9H) ppm. 13C NMR (126 MHz; CD2Cl2, 298 K): δ
140.6 (dd, JCP = 7.7, 2.0 Hz, Cq), 133.7-133.5 (m), 133.4-133.3
(m), 131.0 (br m, Cq), 129.4, 129.3, 128.6-128.3 (m), 127.3 (d,
J
CP = 2.9 Hz, Cq), 125.1, 122.0, 119.7, 114.70, 10.32 (CH3) ppm.
31P{1H}-NMR (202 MHz; CD2Cl2; 298 K): δ 58.70 (br q, J =
195.9 Hz, 1P), -27.68 (br d, J = 197.6 Hz, 3P) ppm.
Tris-N-(3-methyl-2-diisopropylphosphinoindolyl)phosphine (2).
To a solution of diisopropyl(3-methyl-2-indolyl)phosphine (0.51 g,
2.1 mmol) in THF (10 mL) was added n-BuLi (2.5 M in hexanes,
0.83 mL, 2.1 mmol) at -78 °C. The resulting solution was stirred
for 30 min, and phosphorus trichloride (369 μL, 4.2 mmol) was
added. The reaction mixture was stirred overnight allowing to
warm to rt. The resulting solution was concentrated in vacuo, and
the off-white residue was purified by flash SiO2 chromatography
(3% EtOAc inhexanes). Yield:0.36g (68%). 1HNMR(500MHz;
CDCl3; 298 K): δ 7.49 (d, J = 7.7 Hz, 3H), 6.99 (t, J = 7.4 Hz,
3H), 6.69 (t, J = 7.4 Hz, 3H), 6.35 (d, J = 8.5 Hz, 3H), 2.48-2.43
(br m, 3H), 2.39 (s, 9H), 2.23-2.19 (br m, 3H), 1.24 (dd, J = 15.7,
6.9 Hz, 18H), 0.77 (dd, J = 17.2, 6.8 Hz, 9H), -0.06 (dd, J = 13.3,
6.8 Hz, 9H) ppm. 31P{1H}-NMR (202 MHz; CDCl3, 298 K): δ
67.27 (q, J = 219.5 Hz, 1P), -9.92 (d, J = 219.3 Hz, 3P) ppm.
[Cu(2)Cl] (5). Tetraphosphine 2 (50 mg, 0.065 mmol) and
[CuCl(cod)]2 (12 mg, 0.028 mmol) are stirred in CH2Cl2 (2 mL)
for 1 h at rt. The resulting pale yellow solution was concentrated
under reduced pressure, and hexanes (10 mL) were added to
precipitate a pale yellow solid. The solid was collected by
filtration and washed with hexanes (5 mL). Crystals suitable
for X-ray single crystal diffraction were obtained by slow
diffusion of hexane in a CH2Cl2 solution at rt. Yield: 18 mg
(37%). 1H NMR (300 MHz; CD2Cl2; 298 K): δ 7.57 (d, J = 8.1
Hz, 3H), 7.09 (t, J = 7.4 Hz, 3H), 6.72 (t, J = 7.8 Hz, 3H), 6.01
(br d, J = 8.4 Hz, 3H), 2.89 (br s, 3H), 2.47 (s), 1.95 (quintuplet,
J = 7.1 Hz, 3H), 1.43-1.24 (m, 18H), 0.75-0.57 (m, 18H) ppm.
31P{1H}-NMR (121.5 MHz; CD2Cl2; 298 K): δ 59.27 (q, J =
172.2 Hz, 1P), -2.32 (d, J = 173.7 Hz, 3P) ppm.

Article
[Cu(3)Cl] (6). Tetraphosphine 3 (153 mg, 0.16 mmol) and
Inorganic Chemistry, Vol. 49, No. 14, 2010 6507
obtained by slow diffusion of hexane in a CH2Cl2 solution at rt.
Yield: 54 mg (48%). Mp = 220 °C. Anal. Calcd for the CH2Cl2
adduct C64H53Cl3F6N3P5Rh (%): C 57.27, H 3.98, N 3.13;
found: C 57.88, H 4.06, N 3.24. HRMS (FAB) calcd for [M -
PF6]þ C63H51ClN3P4Rh, 1111.1777; found, 1111.1777.
[CuCl(cod)]2 (33 mg, 0.08 mmol) are stirred in CH2Cl2 (8 mL)
for 1 h at rt. The resulting yellow solution was concentrated under
reduced pressure and hexanes (10 mL) were added to precipitate a
yellow solid. The solid was collected by filtration and washed with
hexanes (5 mL). Crystals suitable for X-ray single crystal diffrac-
tion were obtained by slow diffusion of hexane in a CH2Cl2
solution at rt. Yield: 165 mg (98%). 1H NMR (500 MHz; CD2Cl2;
298 K): δ 7.37-7.28 (m, 21H), 7.15 (br m, 12H), 7.01 (t, J = 7.5
Hz, 3H), 6.85 (t, J = 7.6 Hz, 3H), 6.76 (d, J = 8.4 Hz, 3H), 1.96 (s,
9H) ppm. 13C NMR (126 MHz; CD2Cl2, 298 K): δ 140.7 (Cq),
134.3 (d, JCP = 6.3 Hz, Cq), 132.9 (br, Cq), 132.5 (br), 130.0 (br),
128.6 (br), 125.8 (Cq), 123.9, 121.2, 120.0, 115.3, 10.9 (CH3) ppm.
31P NMR (202 MHz; CD2Cl2): δ 40.30 (br d, JPP = 119.7 Hz, 3P),
-63.55 (br m, 1P) ppm.
[Ir(1)Cl] (9). Tetraphosphine 1 (100 mg, 0.10 mmol) and
[Ir(cod)Cl]2 (35 mg, 0.05 mmol) are stirred in CH2Cl2 (4 mL) for
1 h at rt. The resulting orange solution was filtered through Celite
and concentrated under reduced pressure. The solid was washed
with hot EtOH (10 mL) and dried in vacuo. Yield: 78 mg (65%).
Mp = 224 °C (dec.). 1H NMR (500 MHz; CD2Cl2; 298 K):
δ 7.50 (d, J = 7.8 Hz, 3H), 7.21 (t, J = 7.4 Hz, 6H), 7.11-6.93
(m, 30H), 6.87 (d, J = 8.4 Hz, 3H), 1.63 (s, 9H). 13C NMR (126
MHz; CD2Cl2, 298 K): δ138.8-138.7 (m), 134.5 (m), 133.82, 133.6
(m), 131.6-131.4 (m), 129.0, 128.8-128.6 (m), 128.3-128.2 (m),
128.1-127.7 (m), 125.3, 122.5-122.4 (m), 121.2-121.0 (m), 114.4
(m), 10.7 (CH3). 31P{1H}-NMR (202 MHz; CD2Cl2; 298 K): δ
93.15 (q, JPP = 13.7 Hz, 1P), -6.96 (d, JPP = 14.8 Hz, 3P). Anal.
Calcd for C63H51ClIrN3P4 (%): C 62.97, H 4.28, N 3.50; found: C
[Rh(1)Cl] (7).17. Tetraphosphine 1 (800 mg, 0.82 mmol) and
[Rh(cod)Cl]2 (203 mg, 0.41 mmol) are stirred in CH2Cl2 (40 mL)
for 0.5 h at rt. The resulting deep red solution was concentrated
under reduced pressure, and hexanes (100 mL) were added to
precipitate a dark red solid. The solid was collected by filtration
and washed with hexanes (20 mL). Crystals suitable for X-ray
single crystal diffraction were obtained by slow diffusion of hexane
in a THF solution at rt. Yield: 834 mg (91%). Mp = 319 °C. 1H
NMR (500 MHz; CDCl3; 298 K): δ 7.46 (d, J = 7.9 Hz, 3H), 7.29
(br s, 6H), 7.22 (t, J = 7.3 Hz, 3H), 7.18 (t, J = 7.6 Hz, 3H), 7.10 (t,
J= 7.5 Hz, 12H), 7.04 (t, J= 7.3 Hz, 3H), 6.98 (t, J= 7.8 Hz, 3H),
62.36,
H 4.51, N
3.24. HRMS (FAB) calcd for [M]þ
C63H51ClIrN3P4, 1201.2349; found 1201.2345.
[Ir(3)Cl] (10). Tetraphosphine 3 (200 mg, 0.21 mmol) and
[Ir(cod)Cl]2 (70 mg, 0.10 mmol) are stirred in CH2Cl2 (8 mL) for
3 days at rt. The resulting orange suspension was concentrated
under reduced pressure. Et2O (20 mL) was added, and the
orange solid was collected by filtration and dried in vacuo.
Yield: 203 mg (82%). Mp = 272 °C (dec.). 1H NMR (500 MHz;
CD2Cl2, 298 K): δ 7.56 (d, J = 8.0 Hz, 3H), 7.08 (t, J = 7.3 Hz,
3H), 7.05-6.99 (m, 12H), 6.85 (m, 18H), 6.77 (t, J = 7.8 Hz, 3H),
6.38 (d, J = 8.4 Hz, 3H), 2.62 (d, J = 1.4 Hz, 9H). 13C NMR (126
MHz; CD2Cl2, 298 K): δ 139.3-139.2 (m), 138.3-137.9 (m),
136.8 (d, JCP = 9.3 Hz), 129.9 (m), 128.2 (m), 124.2, 122.1, 121.2,
120.9, 116.4, 11.0 (CH3). 31P{1H}-NMR (202 MHz; CD2Cl2, 298
K): δ 53.67 (d, JPP = 17.9 Hz, 3P), 6.38 (q, JPP = 17.9 Hz, 1P).
Anal. Calcd for C63H51ClIrN3P4 (%): C 62.97, H 4.28, N 3.50;
found: C 61.14, H 4.70, N 3.11. HRMS (FAB) calcd for [M þ H]þ
C63H52ClIrN3P4, 1202.2427; found 1202.2412.
6.83 (t, J = 7.6 Hz, 6H), 6.70 (d, J = 8.4 Hz, 3H), 1.59 (s, 9H). 13
C
NMR (126 MHz; CDCl3; 298 K): δ 139.4 (m), 136.1 (m),
134.2-133.9 (m), 130.8-130.5 (m), 128.9 (m), 128.5, 128.2 (m),
127.9, 125.3, 122.1, 120.5, 113.8, 10.3 (CH3). 31P{1H}-NMR (202
MHz; CDCl3; 298 K): δ 131.75 (dq, JPRh = 169.7 Hz, JPP = 30.3
Hz, 1P), 5.74 (dd, JPRh = 138.0 Hz, JPP = 30.2 Hz, 3P). Anal.
Calcd for the CH2Cl2 adduct C64H53Cl3N3P4Rh (%): C 64.20, H
4.46, N 3.51; found: C 64.19, H 4.55, N 3.41. HRMS (FAB) calcd
for [M þ H]þ C63H52ClN3P4Rh, 1112.1855; found, 1112.1849.
[Rh(1)Cl]PF6 ([7]PF6). Complex 7 (80 mg, 0.072 mmol) and
ferrocenium hexafluorophosphate (24 mg, 0.072 mmol) are
stirred in CH2Cl2 (4 mL) for 1 h at rt. The resulting deep purple
solution was concentrated under reduced pressure. The dark
purple residue was washed with Et2O (3 ꢀ 10 mL) and dried in
vacuo. Yield: 77 mg (85%). Mp = 190 °C. Anal. Calcd for the
CH2Cl2 adduct C64H53Cl3F6N3P5Rh (%): C 57.27, H 3.98,
N 3.13; found: C 57.32, H 4.68, N 3.01. HRMS (FAB) calcd
for [M - PF6]þ C63H51ClN3P4Rh, 1111.1777; found, 1111.1786.
[Rh(3)Cl] (8).18. Tetraphosphine 3 (400 mg, 0.41 mmol) and
[Rh(cod)Cl]2 (101 mg, 0.21 mmol) are stirred in CH2Cl2 (20 mL)
for 0.5 h at rt. The resulting deep red solution was concentrated
under reduced pressure and hexanes (40 mL) were added to
precipitate a burgundy-colored solid. The solid was collected by
filtration and washed with hexanes (2 ꢀ 10 mL). Crystals suitable
for X-ray single crystal diffraction wereobtained by slow diffusion
of hexane in a CH2Cl2 solution at rt. Yield: 378 mg (83%). Mp =
303 °C (dec.). 1H NMR (500 MHz; CDCl3; 298 K): δ 7.55 (d, J =
8.0 Hz, 3H), 7.05-7.02 (m, 15H), 6.82-6.77 (m, 21H), 6.42 (d,
J = 8.5 Hz, 3H), 2.55 (s, 9H). 13C NMR (126 MHz; CDCl3; 298
K): δ 139.7-139.6 (m), 136.8-136.7 (m), 136.1 (d, JCP = 8.7 Hz),
129.8 (m), 128.1 (d, JCP = 4.9 Hz), 124.1, 122.3 (m), 121.0, 120.3,
116.3, 10.5 (CH3). 31P{1H}-NMR (202 MHz; CDCl3; 298 K): δ
[Rh(1)(nbd)]BF4 (11). Tetraphosphine 1 (200 mg, 0.20 mmol)
and [Rh(nbd)2]BF4 (77 mg, 0.20 mmol) are stirred in CH2Cl2
(10 mL) for 1 h at rt. The resulting orange-red solution was
concentrated under reduced pressure, and hexanes (10 mL) were
added to precipitate an orange-red solid. The solid was collected
by filtration and washed with hexanes (10 mL). Yield: 260 mg
(99%). 1H NMR (500 MHz; CD2Cl2; 298 K): δ 7.57 (d, J = 7.9
Hz, 3H), 7.39 (t, J = 7.7 Hz, 3H), 7.34 (t, J = 7.9 Hz, 6H), 7.20
(t, J = 7.6 Hz, 6H), 7.14 (t, J = 7.6 Hz, 9H), 7.08 (t, J = 7.5 Hz,
6H), 6.81 (s, 6H), 6.58 (d, J = 8.4 Hz, 3H), 6.34-6.32 (m, 1H),
6.26-6.25 (m, 1H), 5.37 (s, 1H), 4.55 (s, 1H), 2.76 (s, 1H), 2.38 (s,
1H), 1.28 (s, 9H), 1.05 (d, J = 7.6 Hz, 1H), 1.00 (d, J = 7.6 Hz,
1H) ppm. 13C NMR (126 MHz; CD2Cl2, 298 K): δ 139.8, 136.5,
132.5 (m), 130.6 (m), 130.5, 130.1, 129.5 (m), 129.1 (m), 127.5,
127.4, 127.2, 123.7, 121.7, 113.5, 49.3, 48.5, 9.61 (CH3) ppm.
31P{1H}-NMR (202 MHz; CD2Cl2; 298 K): δ 132.36 (dq,
JRhP = 137.8 Hz, JPP = 36.7 Hz, 1P), 6.43 (dd, JRhP = 129.5
Hz, JPP = 36.3 Hz, 3P) ppm. Anal. Calcd for the CH2Cl2 adduct
C71H61BCl2F4N3P4Rh (%): C 63.60, H 4.59, N 3.13; found: C
64.00, H 4.98, N 3.00. HRMS (FAB) calcd for [M - nbd -
BF4]þ C63H51N3P4Rh, 1076.2089; found 1076.2081.
73.95(dd, JPRh = 158.8 Hz, JPP = 26.9 Hz, 3P), 43.49 (dq, JPRh
=
[Rh(1)(CO)]BF4 (12). In a stainless steel autoclave, a solution
of complex 11 (100 mg, 0.08 mmol) in CH2Cl2 (3 mL) was stirred
overnight under 25 bar CO at rt. The resulting orange solution
was concentrated in vacuo, and the residue was washed with
hexanes (2 ꢀ 5 mL) to give a light orange powder. Crystals
suitable for X-ray single crystal diffraction were obtained by
slow diffusion of hexane in a CH2Cl2 solution at rt. Yield: 49 mg
(51%). FT-IR: νCO = 2000 cm-1. 1H NMR (500 MHz; CD2Cl2;
298 K): δ 7.58 (d, J = 8.0 Hz, 3H), 7.46 (t, J = 7.3 Hz, 3H), 7.35
(t, J = 7.6 Hz, 3H), 7.28 (t, J = 7.5 Hz, 6H), 7.22-7.15 (m,
12H), 6.95-6.87 (m, 12H), 6.73 (d, J = 8.4 Hz, 3H), 1.60 (s, 9H)
108.2 Hz, JPP = 26.8 Hz, 1P). Anal. Calcd for the CH2Cl2 adduct
C64H53Cl3N3P4Rh (%): C 64.20, H 4.46, N 3.51; found: C 64.22,
H 4.51, N 3.48. HRMS (FAB) calcd for [M]þ C63H51ClN3P4Rh,
1111.1777; found, 1111.1771.
[Rh(3)Cl]PF6 ([8]PF6).18. Complex 8 (100 mg, 0.09 mmol) and
ferrocenium hexafluorophosphate (30 mg, 0.09 mmol) are stir-
red in CH2Cl2 (5 mL) for 1 h at rt. The resulting deep purple
solution was concentrated under reduced pressure. The dark
purple residue was washed with Et2O (3 ꢀ 10 mL) and dried in
vacuo. Crystals suitable for X-ray single crystal diffraction were

6508 Inorganic Chemistry, Vol. 49, No. 14, 2010
Wassenaar et al.
ppm. 13C NMR (126 MHz; CD2Cl2; 298 K): δ 139.7 (m), 136.7
(m), 135.2-134.7 (m), 133.8-133.5 (m), 133.2 (m), 131.4, 130.6
(m), 130.2, 129.4 (m), 129.2 (m), 128.6 (d, JCP = 14.9 Hz), 127.5,
124.2, 121.9, 113.8, 10.2 (CH3) ppm. 31P{1H}-NMR (202 MHz;
CD2Cl2; 298 K): δ 129.63 (dq, JRhP = 115.0 Hz, JPP = 38.8 Hz,
1P), 16.26 (dd, J = 126.3 Hz, JPP = 39.1 Hz, 3P) ppm. Anal.
Calcd for C64H51BF4N3OP4Rh (%): C 64.50, H 4.31, N 3.53;
found: C 63.24, H 4.77, N 3.14. HRMS (FAB) calcd for [M -
BF4]þ C64H51N3OP4Rh, 1104.2038; found 1104.2043.
CuI-Catalyzed Cyclopropanation. AgOTf (1.3 mg, 0.005
mmol) and the corresponding copper complex (0.005 mmol) were
stirred at rt in dry CH2Cl2 (5 mL) for 10 min. To this mixture,
styrene (0.458 mL, 4.0 mmol) and EDA (0.053 mL, 0.5 mmol)
were added, and the reaction mixture was stirred for 18 h at rt.
Evaporation of the solvent gave the crude product, which was
analyzed by 1H NMR to determine conversion and selectivity.
EPR Spectroscopy. Experimental X-band EPR spectra were
recorded on a Bruker EMX spectrometer equipped with a He
temperature control cryostat system (Oxford Instruments). The
spectra were simulated by iteration of the anisotropic g values,
(super)hyperfine coupling constants and line widths.
[Rh(1)(H)Cl]PF6 (13)(PF6). Complex [7]PF6 (19 mg, 0.02 mmol)
was charged to a 5 mm high-pressure NMR-tube and dissolved in
CD2Cl2. The tube was put under 5 bar of H2, and the reaction
followed by NMR spectroscopy. Quantitative conversion to
complex 13 was observed after 6 days. Unambiguous character-
ization of this compound as [RhIII(1)(H)Cl]PF6 was established
by comparison with an authentic sample (vide infra, different
anion). [Rh(1)(H)Cl]OTf ((13)(OTf)) was prepared by reaction of
7 (150 mg, 0.14 mmol) with1 equivof HOTf (22mg, 0.15 mmol) in
CH2Cl2 (8 mL) for 1 h. The solvent was evaporated in vacuo, and
the residue was washed with pentanes. After drying in vacuo, a
dark-purple solid was obtained. Yield: 151 mg (89%). 1H NMR
(500 MHz; CD2Cl2; 298 K): δ 7.84 (d, J = 7.9 Hz, 1H), 7.67-7.55
(m, 6H), 7.54-7.22 (m, 18H), 7.20-7.16 (m, 2H), 7.12 (td, J =
7.8, 2.7 Hz, 2H), 7.04 (t, J = 7.2 Hz, 1H), 6.91 (d, J = 8.5 Hz, 1H),
6.73-6.63 (m, 6H), 6.53 (dd, J = 11.5, 8.0 Hz, 2H), 6.37 (d, J =
8.5 Hz, 1H), 6.23 (d, J = 8.4 Hz, 1H), 5.97 (br s, 1H), 2.03 (s, 3H),
DFT Geometry Optimizations and Spectral Parameter Calcu-
lations. The geometry optimizations were carried out with the
Turbomole program44a coupled to the PQS Baker optimizer.45
Geometries were fully optimized as minima at the BP8646 level
using the Turbomole SV(P) basis set44c,d on all atoms. EPR
parameters47 were calculated with the ADF48 program system
using the BP8646 functional with the ZORA/TZP basis set supplied
with the program (all electron, core double-ζ, valence triple-ζ
polarized basis set on all atoms), using the coordinates from the
structures optimized in Turbomole as input. Orbital and spin
density plots were generated with Molden.49 Calculated UV-vis
spectra based on TD-DFT calculations were obtained with Orca50
at the ri-DFT, BP86, SV(P)-SV/J level, using the Tamm-Dancoff
approximation (TDA) and COSMO (CH2Cl2) solvent corrections.
X-ray crystallography. All data and details are compiled in the
Supporting Information.
1.70 (s, 3H), 1.61 (s, 3H), -6.49 (ddt, JHPtrans = 167.6 Hz, JHPcis
=
19.1 Hz, JHRh = 9.6 Hz, 1H) ppm. 13C NMR (126 MHz; CD2Cl2;
298 K): δ 140.5 (d, JCP = 5.4 Hz, Cq), 140.4 (d, JCP = 5.0 Hz, Cq),
139.82 (d, JCP = 4.7 Hz, Cq), 139.76 (d, JCP = 4.9 Hz, Cq), 139.3
(m, Cq), 137.40 (d, JCP = 3.8 Hz, Cq), 137.35 (d, JCP = 3.5 Hz,
Cq), 136.7 (d, JCP = 3.3 Hz, Cq), 136.6 (d, J = 3.8 Hz, Cq), 135.6,
135.5, 135.21 (d, JCP = 4.4 Hz, Cq), 135.15 (d, JCP = 4.3 Hz, Cq),
135.0 (m, Cq), 134.8 (Cq), 134.5 (Cq), 134.3 (Cq), 134.0 (Cq), 133.6,
133.5, 133.1 (d, JCP = 2.1 Hz), 133.0 (d, JCP = 2.1Hz), 132.7 (Cq),
132.5 (d, JCP = 2.7 Hz), 132.3 (Cq), 132.2 (Cq), 131.9(d, JCP = 2.3
Hz), 131.8 (d, JCP = 3.1 Hz), 131.7, 131.6, 131.31 (d, JCP = 2.4
Hz), 131.24, 131.16 (d, JCP = 1.3 Hz), 130.9, 130.8, 130.5 (d,
Acknowledgment. This research was supported by the
National Research School Combination on Catalysis (NRSC-C),
the University of Amsterdam and the Marie-Curie International
Training Network (R)EvCAT funded by the EU. J.I.v.d.V.
is grateful to the Dutch Research Council-Chemical Sciences
(NWO-CW) for a VENI Innovative Research grant. We thank
J. W. H. Peeters for measuring high resolution mass spectra,
Dr. D. G. H. Hetterscheid for help with NMR spectroscopy,
T. Mahabiersing and W.I. Dzik for assistance with cyclic voltam-
metry measurements, Dr. Martin Lutz (Univ. Utrecht) for solving
the molecular structure of complex 6 by X-ray crystallography,
and Prof. F. Neese (Univ. Bonn, Germany) for a copy of his EPR
simulation program.
JCP = 2.4 Hz), 130.3 (Cq), 130.0 (Cq), 129.7, 129.6, 129.5, 129.4,
129.3, 129.2 (d, JCP = 3.1 Hz), 128.9, 128.8, 128.0 (d, JCP = 4.1
Hz), 127.9, 125.9 (Cq), 125.8, 125.6 (Cq), 125.5, 125.4 (m, Cq),
124.9, 124.5, 124.1 (m, Cq), 124.0 (d, JCP = 9.4 Hz, Cq), 123.7 (d,
J
CP = 9.1 Hz, Cq), 122.5, 122.3, 122.1, 116.1, 114.0 (d, JCP = 3.0
Hz), 12.0 (CH3), 11.9 (CH3), 10.6 (CH3) ppm. 31P{1H}-NMR
Supporting Information Available: NMR spectra of complex
13, molecular structures of ligands 1 and 3 and complex 7,
experimental details for reactivity of 7 toward acids, UV-vis
spectra and TD-DFT simulations of 7 and [7]PF6, X-ray crystal-
lographic details and crystallographic information files (CIF).
This material is available free of charge via the Internet at http://
pubs.acs.org.
(202 MHz; CD2Cl2; 298 K): δ 115.71 (m, 1P), 20.63 (dddd,
JPPtrans = 371.3 Hz, JPRh = 88.1 Hz, JPPcis = 26.6 Hz, JPPcis =
17.8 Hz, 1P), 15.00 (ddt, JPPtrans = 371.2 Hz, JPRh =97.3Hz,JPPcis =
17.2 Hz, 1P), -1.69 (m, 1P) ppm. The JPP and JRhP coupling
constants are disturbed by the JHP coupling from the hydride,
which is not entirely decoupled.
€
€
(44) (a) Ahlrichs, R.; Bar, M.; Baron, H.-P.; Bauernschmitt, R.; Bocker,
(47) Some references and reviews on DFT approaches to EPR parameters:
(a) van Lenthe, E.; van der Avoird, A.; Wormer, P. E. S. J. Chem. Phys. 1997,
107, 2488. (b) van Lenthe, E.; van der Avoird, A.; Wormer, P. E. S. J. Chem. Phys.
1998, 108, 4783. (c) Neese, F. Curr. Opin. Chem. Biol. 2003, 7, 125. (d) Neese, F.;
Solomon, E. In Magnetoscience - From Molecules to Materials; Miller, J. S.,
Drillon, M., Eds.; Wiley: New York, 2003; pp 345-466; (e) Peng, G.; Nichols, J.;
McCullough, E. A.; Spence, J. Inorg. Chem. 1994, 33, 2857–2864.
(48) (a) Baerends, E. J.; Ellis, D. E.; Ros, P. Chem. Phys. 1973, 2, 41. (b)
Versluis, L.; Ziegler, T. J. Chem. Phys. 1988, 88, 322. (c) te Velde, G.; Baerends,
E. J. J. Comput. Phys. 1992, 99, 84. (d) Fonseca Guerra, C.; Snijders, J. G.; te
Velde, G.; Baerends, E. J. Theor. Chem. Acc. 1998, 99, 391.
€
S.; Ehrig, M.; Eichkorn, K.; Elliott, S.; Furche, F.; Haase, F.; Haser, M.;
Hattig, C.; Horn, H.; Huber, C.; Huniar, U.; Kattannek, M.; Kohn, A.;
€
€
€
€
€
Kolmel, C.; Kollwitz, M.; May, K.; Ochsenfeld, C.; Ohm, H.; Schafer, A.;
Schneider, U.; Treutler, O.; Tsereteli, K.; Unterreiner, B.; von Arnim, M.;
Weigend, F.; Weis, P.; Weiss, H. Turbomole, Version 5; Theoretical Chemistry
Group, University of Karlsruhe: Karlsruhe, Germany, 2002. (b) Treutler, O.;
Ahlrichs, R. J. Chem. Phys. 1995, 102, 346. (c) Turbomole basis set library,
Turbomole Version 5, see : (d) Schafer, A.; Horn, H.; Ahlrichs, R. J. Chem. Phys.
1992, 97, 2571. (e) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H.
Theor. Chim. Acta 1990, 77, 123. (f ) Schafer, A.; Huber, C.; Ahlrichs, R. J.
€
€
(49) Schaftenaar, G.; Noordik, J. H. J. Comput. Aided Mol. Des. 2000, 14,
123–134.
(50) ORCA;an ab initio, density functional and semiempirical program
package, Version 2.7.0; Written by Neese, F. with contributions from Becker, U.;
Ganyushin, D.; Liakos, D.; Kossmann, S.; Petrenko, T.; Riplinger, C.; Wennmohs,
Chem. Phys. 1994, 100, 5829.
(45) (a) PQS, version 2.4; Parallel Quantum Solutions: Fayetteville, AK, 2001
(the Baker optimizer is available separately from PQS upon request); (b) Baker, J.
J. Comput. Chem. 1986, 7, 385.
(46) (a) Becke, A. D. Phys. Rev. A 1988, 38, 3098–3100. (b) Perdew, J. P.
Phys. Rev. B 1986, 33, 8822–8824.
€
F. Universitat Bonn: Bonn, Germany, 2009.