A homologous series of cobalt, rhodium, and iridium metalloradicals

Article abstract of DOI:10.1021/ic202079r

We herein present a series of d⁷ trimethylphosphine complexes of group 9 metals that are chelated by the tripodal tetradentate tris(phosphino)silyl ligand [SiP^iPr₃]H ([SiP ^iPr₃] = (2-iPr₂PC₆H ₄)₃Si^-). Both electron paramagnetic resonance (EPR) simulations and density functional theory (DFT) calculations indicate largely metalloradical character. These complexes provide a rare opportunity to compare the properties between the low-valent metalloradicals of the second- and third-row transition metals with the corresponding first-row analogues.

Full text of DOI:10.1021/ic202079r

Communication
pubs.acs.org/IC
A Homologous Series of Cobalt, Rhodium, and Iridium
Metalloradicals
Ayumi Takaoka and Jonas C. Peters*
Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, California 91125, United States
S
* Supporting Information
Scheme 1. Synthesis of Divalent Group 9 Complexes
ABSTRACT: We herein present a series of d7 trimethyl-
phosphine complexes of group 9 metals that are chelated
by the tripodal tetradentate tris(phosphino)silyl ligand
[SiPiPr3]H ([SiPiPr3] = (2-iPr2PC6H4)3Si−). Both electron
paramagnetic resonance (EPR) simulations and density
functional theory (DFT) calculations indicate largely
metalloradical character. These complexes provide a rare
opportunity to compare the properties between the low-
valent metalloradicals of the second- and third-row
transition metals with the corresponding first-row
analogues.
ow-valent metalloradicals of the second- and third-row late
L
transition metals are, in general, reactive species that have
often necessitated in situ characterization.1 There are a small
1
number of well-defined examples of these S =
/ metal-
2
Si−H bond activation by the addition of [SiPiPr3]H to
[Rh(COD)Cl]2 in 87% yield. As previously described for 6,6b
centered radicals, however, that point to their interesting
spectroscopic properties and reactivity patterns.1,2 Wayland’s
classic studies on rhodium(II) porphyrin complexes nicely
illustrate the latter point.3 The relative instability of these
second- and third-row metalloradicals in comparison with their
first-row congeners constitutes an interesting dichotomy in the
chemistry of late transition metals.4 Few studies, however, have
compared the properties of late metalloradicals within a group
that possess similar geometries and ancillary ligands.5
complex 5 features a labile N2 ligand (νN = 2159 cm−1).
2
For the rhodium and iridium systems, the addition of excess
PMe3 to complexes 5 and 6 leads to clean substitution to afford
the yellow PMe3 complexes [SiPiPr3]M(PMe3) [M = Rh (8), Ir
(9)]. Complexes 8 and 9 reveal reversible oxidation waves at
−0.78 and −0.76 V (vs Fc/Fc+, THF), respectively.
Accordingly, the oxidation of 8 and 9 with FcBArF [Fc =
4
Fe(C5H5)2] results in color changes to blue and purple,
respectively, and affords the desired 17 e−, S = 1/2 complexes 2
(66%) and 3 (88%).
We have recently employed a tripodal, tris(phosphino)silyl
ligand, [SiPiPr3]H ([SiPiPr3] = (2-iPr2PC6H4)3Si−),6 to stabilize
a number of group 8 metalloradicals, including unusual
examples of mononuclear ruthenium(I) and osmium(I)
complexes.7 These metalloradicals included a series of
dinitrogen complexes of iron(I), ruthenium(I), and osmium-
(I).6,7 We herein report on the synthesis and thorough
characterization of a related series of isoelectronic PMe3 adduct
complexes of the group 9 metals {[SiPiPr3]M(PMe3)}BArF4 [M
In contrast to 2 and 3, the cobalt metalloradical 1 is
synthesized by the addition of FcBArF4 to a solution containing
1
4 and excess PMe3, which yields orange, S = /2, complex 1 in
58% after workup. Interestingly, the reduction of isolated 1 by
CoCp*2 under an N2 atmosphere cleanly regenerates the CoI−
N2 adduct 4 with quantitative loss of PMe3. Further, complex 4
exhibits no tendency to bind PMe3 under an atmosphere of N2.
Hence, the apparent stronger preference of CoI for N2 over
PMe3 in comparison to the related RhI and IrI fragments
appears to be thermodynamic rather than kinetic in origin.
The solid-state structures of 1−3 have been obtained
through X-ray diffraction studies (Figure 1 and SI). The
geometries about the metal centers are similar, exhibiting
distorted trigonal-bipyramidal (TBP) geometries [τ = 0.81 (1),
0.75 (2), 0.73 (3)].8 This correspondence allows for a
= Co (1), Rh (2), Ir (3); BArF = tetrakis[3,5-bis-
4
(trifluoromethyl)phenyl]borate]. These complexes constitute
a rare instance wherein a series of metalloradicals within a
group can be isolated and for which their ancillary ligands,
oxidation states, spin states, and geometries are conserved.
Entry to the desired d7, group 9 complexes begins with the
dinitrogen complexes [SiPiPr3]M(N2) [M = Co (4), Rh (5), Ir
(6); Scheme 1]. While complexes 4 and 6 have been
reported,6b complex 5 has not been previously synthesized.
Briefly, 5 is prepared through dehydrodehalogenation of a
hydrido chloride complex, [SiPiPr3]Rh(H)(Cl) (7), with
MeMgBr in 97% yield. Complex 7, in turn, is prepared via
Received: September 23, 2011
Published: December 1, 2011
© 2011 American Chemical Society
16
dx.doi.org/10.1021/ic202079r | Inorg. Chem. 2012, 51, 16−18

Inorganic Chemistry
Communication
Figure 1. Solid-state structures of {[SiPiPr3]Co(PMe3)}{BArF } (left)
4
and {[SiPiPr3]Ir(PMe3)}{OTf} (right) at 50% probability. BArF and
4
OTf anions, H atoms, and solvent molecules are removed for clarity.
For 2, see SI.
comparison of their chemical and spectroscopic properties.
Complexes 1−3 each exhibit one P−M−P angle [129.54(4)°
(1), 131.99(3)° (2), 133.51(4)° (3)] that is substantially larger
than the other two, a feature that was also observed in the solid-
state structures of the Ru−N2 and Os−N2 metalloradicals,
7
[SiPiPr ]M(N2) (M = Ru (11), Os(12)). The doubly
3
2
degenerate E ground state of an idealized TBP structure,
2
2
whereby the dxy/dx −y orbitals are triply occupied, is subject to
Jahn−Teller distortion, consistent with the distorted structural
parameters of 1−3.
To assess the metalloradical character of complexes 1−3,
their EPR spectra were measured at X-band frequency.
Deviations of the isotropic g value from the free-electron
value of 2.0023 and the g anisotropy in a frozen solution, Δg
(Δg = gmax − gmin, where gmax and gmin are the largest and
smallest g tensors), have been used as crude indicators in
assessing the metalloradical character; large deviations from g =
2.0023 at room temperature and higher values of Δg in a frozen
solution typically point to predominant spin on the metal.1 The
room temperature EPR spectra of 2 and 3 are rather featureless
(see the SI). For 1 a signal could not be observed at room
temperature, likely because of the rapid relaxation induced by
the metal center. Complexes 2 and 3 exhibit spectra
reminiscent of 11 and 12, showing broad doublet signals.
The splitting pattern indicates strong coupling to one P atom,
while coupling to the other P atoms is smaller and unresolved.
The observation of asymmetry in the P atoms even at room
temperature is due to the faster time scale of the EPR
experiment relative to the NMR experiment, where an averaged
3-fold symmetry is suggested. The g values for 2 and 3, which
are 2.100 and 2.145, respectively, are similar to the values of 11
(2.078) and 12 (2.147).
The 77 K X-band EPR spectra taken in 2-MeTHF are shown
in Figure 2 and are more revealing. All three spectra are
rhombic, with significant anisotropy (Table 1). The Δg values
for 1−3 are 0.61, 0.18, and 0.33 and, together with the room
temperature isotropic g values, suggest significant metalloradical
character. While the spectrum of 1 is broad and thus does not
show resolvable hyperfine coupling with either the Co or P
atoms, the spectra of 2 and 3 exhibit sharp splitting. Both
spectra have been simulated by assigning a large hyperfine
coupling to one P atom, with smaller coupling to either the P,
Rh, or Ir atoms. Large coupling to only one P atom is also seen
in the EPR spectra of 11 and 12 at 77 K and is ascribed to
coupling to the P atom opposite the largest P−M−P angle that
is observed in the solid-state structure. Similar characteristics
have been previously proposed for spectroscopically charac-
Figure 2. Spin-density plots (left) and 77 K X-band EPR in 2-MeTHF
(right) of 1 (top), 2 (center), and 3 (bottom). The spin-density plots
are shown looking down the Me3P−M−Si axis, with the P atom
opposite to the largest P−M−P angle at the top. The lower curves in
each EPR spectra represent simulations. See the SI for calculation
details and full simulation parameters.
a
Table 1. EPR Parameters for Complexes 1−3, 11, and 12
b
b
Co(1)
Rh(2)
Ir(3)
Ru(11)
Os(12)
gx
2.600
2.080
1.990
0.61
2.205
2.087
2.025
0.18
360
2.300
2.170
1.975
0.33
370
2.175
2.075
2.009
0.17
220
2.290
2.200
1.978
0.31
190
gy
gz
Δg
A(P)x
A(P)y
A(P)z
N/A
N/A
N/A
430
430
230
190
550
500
250
230
a
Hyperfine coupling constants are in megahertz and represent values
for the largest coupled P atom. For the full set of experimental and
b
simulation parameters, see the SI. Parameters from ref 7.
terized rhodium(II) complexes featuring poly(phosphine)
ligands.9 This large coupling is consistent with the doublet
resonance observed at room temperature. The magnitude of
this hyperfine coupling, however, is roughly 2-fold greater for 2
and 3 relative to 11 and 12 (Table 1). Note that, although
complexes 1−3 are metalloradical in character, large hyperfine
coupling constants for the P atoms are observed in 2 and 3
because of the large relative gyromagnetic ratio of P compared
to Rh (P:Rh = −12.9) and Ir (P:191Ir = 22.6; P:193Ir = 20.9).
The conclusions from the EPR data are corroborated by
DFT calculations (Table 2). These calculations place Mulliken
spin densities of 1.17,10 0.74, and 0.73 e− at the metal centers
17
dx.doi.org/10.1021/ic202079r | Inorg. Chem. 2012, 51, 16−18

Inorganic Chemistry
Communication
X-ray crystallographic data in CIF format, synthetic and
spectroscopic details and data, computational details, and
solid-state structures. This material is available free of charge via
the Internet at http://pubs.acs.org.
Table 2. Mulliken Spin Densities Obtained from DFT
Calculations
a
b
b
Co(1)
Rh(2)
Ir(3)
Ru(11)
Os(12)
M
1.167
−0.042
−0.032
−0.025
0.746
0.233
0.732
0.198
0.836
0.135
0.786
0.129
P(total)
Pmax
AUTHOR INFORMATION
Corresponding Author
*E-mail: jpeters@caltech.edu.
■
0.170
0.161
0.086
0.073
PPMe
−0.009
−0.009
−0.009
−0.005
3
a
P(total) represents the total spin density from the P atoms of the
[SiPiPr3] scaffold. Pmax represents the values from the phosphine
ACKNOWLEDGMENTS
■
b
possessing the greatest spin density. Values from ref 7.
This work was supported by the NSF (Grant CHE-0750234).
Charlene Tsay and Larry Henling are acknowledged for
crystallographic assistance.
for 1−3, respectively. While small in 1, delocalization of the
spin density onto the phosphines is evident for 2 and 3, with
values of 0.23 and 0.19 e− distributed among the P atoms of the
[SiPiPr3] scaffold. The greater spin delocalization for 2 and 3
relative to 1 is likely due to the greater covalency of the M−P
bonds in the former. In contrast, the apical PMe3 P atom
possesses negligible spin for all three complexes. The small
degree of delocalization onto the phosphines in 1 may also
explain its featureless 77 K spectrum. Importantly, one P atom
in 2 and 3 possesses a notably greater value (0.17 e− for 2 and
0.16 e− for 3) relative to the other two P atoms. This P atom
lies opposite the largest P−M−P angle in the equatorial plane
of these complexes, and this observation is consistent with the
EPR simulations that assign a large hyperfine coupling to this
atom. The numbers are roughly double the value observed for
the P atom with the largest spin densities in 11 (0.09 e−) and
12 (0.07e−) and suggest a greater spin delocalization for the
group 9 complexes, in agreement with the EPR parameters
(Table 1). Thus, both the EPR simulations and DFT
calculations are qualitatively consistent and point to the
metalloradical character for 1−3, with a greater degree of
spin leakage for the second- and third-row derivatives 2 and 3.
The frontier orbitals of complexes 1−3 are also of interest.
For all three complexes, the LUMO is ligand-based and the
REFERENCES
■
(1) de Bruin, B.; Hetterscheid, D. G. H.; Koekkoek, A. J. J.;
Grutzmacher, H. Prog. Inorg. Chem. 2007, 55, 247.
(2) Cui, W.; Li, S.; Wayland, B. B. J. Organomet. Chem. 2007, 692,
̈
3198.
(3) (a) Sherry, A. E.; Wayland, B. B. J. Am. Chem. Soc. 1990, 112,
1259. (b) Wayland, B. B.; Ba, S.; Sherry, A. E. J. Am. Chem. Soc. 1991,
113, 5305.
(4) Poli, R. Angew. Chem., Int. Ed. 2010, 49, 2. (b) Wayland, B. B.; Ba,
S.; Sherry, A. E. J. Am. Chem. Soc. 1991, 113, 5305.
(5) (a) Palmer, J. H.; Mahammed, A.; Lancaster, K. M.; Gross, Z.;
Gray, H. B. Inorg. Chem. 2009, 48, 9308. (b) Zhai, H.; Bunn, A.;
Wayland, B. Chem. Commun. 2001, 1294. (c) Deblon, S.; Liesum, L.;
Harmer, J.; Schonberg, H.; Schweiger, A.; Grutzmacher, H. Chem.
̈
̈
Eur. J. 2002, 8, 601.
(6) (a) Mankad, N. P.; Whited, M. T.; Peters, J. C. Angew. Chem., Int.
Ed. 2007, 129, 5768. (b) Whited, M. T.; Mankad, N. P.; Lee, Y.;
Oblad, P. F.; Peters, J. C. Inorg. Chem. 2009, 48, 2507.
(7) Takaoka, A.; Gerber, L. C. H.; Peters, J. C. Angew. Chem., Int. Ed.
2010, 49, 4088.
(8) τ = (b − a)/60, where b and a represent the two largest angles.
See: Addison, A. W.; Rao, T. N.; Van Rijn, J. J.; Verschoor, G. C. J.
Chem. Soc., Dalton Trans. 1984, 1349.
(9) Bianchini, C.; Meli, A.; Peruzzini, M.; Vacca, A. Organometallics
1990, 9, 360.
2
2
SOMO and SOMO−1 are of dxy/dx −y parentage (see the SI).
While the LUMO and SOMO energy difference remains
relatively constant, the SOMO and SOMO−1 energy difference
increases from 1 to 3, from 5.7 to 16.9 kcal/mol. Observation
of the largest g anisotropy in complex 1, despite the smaller
spin−orbit coupling constant for Co relative to Rh and Ir, is
thus not only due to the greater spin density on the metal
center, as suggested by DFT calculations, but also of greater
admixture of the SOMO with filled orbitals.
(10) Some negative spin due to spin polarization is observed on the
silicon (−0.09 e−).
(11) Representative examples: (a) Danopoulos, A. A.; Wilkinson, G.;
Hussain-Bates, B.; Hursthouse, M. B. J. Chem. Soc., Dalton Trans. 1992,
3165. (b) García, M. P.; Jimen
Organometallics 1993, 12, 4660. (c) García, M. P.; Jimen
́
ez, M. V.; Oro, L. A.; Lahoz, F. J.
́
ez, M. V.; Oro,
L. A.; Lahoz, F. J.; Casas, J. M.; Alonso, P. J. Organometallics 1993, 12,
3257. (d) Dunbar, K. R.; Haefner, S. C. Organometallics 1992, 11,
1431. (e) Connely, N. G.; Emslie, D. J. H.; Metz, B.; Orpen, A. G.;
Quayle, M. J. Chem. Commun. 1996, 2289. (f) Dzik, W. I.; Arruga, L.
F.; Siegler, M. A.; Spek, A. L.; Reek, J. N. H.; de Bruin, B.
Organometallics 2011, 30, 1902. (g) Connely, N. G.; Emslie, D. J. H.;
Metz, B.; Orpen, A. G.; Quayle, M. J. Chem. Commun. 1996, 2289.
In conclusion, a series of d7 complexes of group 9 metals has
been synthesized and thoroughly characterized. The electronic
structures of these complexes have been probed through EPR
spectroscopy and DFT calculations, and these results suggest
metalloradical character. Comparison of the complexes within
the series indicates greater spin delocalization onto the
phosphines for Rh and Ir relative to Co. Further, a comparison
of the rhodium and iridium complexes, 2 and 3, with their
isoelectronic group 8 analogues, complexes 11 and 12, points
to similar electronic structures for the two sets of complexes
but with increased spin delocalization onto the phosphines for
2 and 3. The degree of covalency observed in the M−P bonds
in complexes 2, 3, 11, and 12 may explain the unusual stability
of these second- and third-row low-valent metalloradicals, for
which few are isolable and structurally characterized.11
ASSOCIATED CONTENT
■
S
* Supporting Information
18
dx.doi.org/10.1021/ic202079r | Inorg. Chem. 2012, 51, 16−18