Three-coordinate late transition metal fluorinated alkoxide complexes

Article abstract of DOI:10.1039/b911758j

Homoleptic fluorinated alkoxide complexes have been prepared from KOC ₄F₉, 1, via salt metathesis routes. One four-coordinate K{K(18C6)}[Co(OC₄F₉)₄], 2, and four three-coordinate complexes: {K(18C6)}[Fe(OC₄F₉) ₃], 3, {K(18C6)}[Co(OC₄F₉)₃], 4, {K(18C6)}[Cu(OC₄F₉)₃], 5, and {K(18C6)}[Zn(OC₄F₉)₃], 6, have been prepared and all except 5 have been characterized crystallographically. Compounds 3, 4, and 6 are very rare examples of monomeric, trigonal alkoxide complexes. All compounds have been characterized with UV-vis and IR spectroscopy, solution magnetic susceptibility, and elemental analysis. In solution, compound 2 exists in an equilibrium with 4 and 1, which has been probed with cyclic voltammetry, supporting energetically different Co²⁺/Co³⁺ potentials in the three-coordinate (E_p,a = ～1.2 V vs Fc/Fc⁺) and four-coordinate (E_p,a = ～0.9 V) geometries. The ligand field engendered by the perfluoro-t-butoxide ligand has been studied with DFT calculations on 4 and the hypothetical [Co(OC₄H₉) ₃]^- as well as the previously reported [Co(mes) ₃]^- and [Co{N(TMS)₂}₃]^- showing significant π-type interactions in the xy plane as well as above and below for the two alkoxide species. The Royal Society of Chemistry.

Full text of DOI:10.1039/b911758j

View Article Online / Journal Homepage / Table of Contents for this issue
This article is published as part of the Dalton Transactions themed issue entitled:
New Talent
Showcasing the strength of research being carried out by
tomorrow's leaders in the field of inorganic chemistry
Guest Editor Polly Arnold
University of Edinburgh, UK
Published in issue 2, 2010 of Dalton Transactions
Image reproduced with permission of Mark Muldoon
Articles in the issue include:
PERSPECTIVES:
Modern multiphase catalysis: new developments in the separation of homogeneous catalysts
Mark J. Muldoon, Dalton Trans., 2010, DOI: 10.1039/b916861n
Probing bioinorganic chemistry processes in the bloodstream to gain new insights into the origin of
human diseases
Elham Zeini Jahromi and Jürgen Gailer, Dalton Trans., 2010, DOI: 10.1039/b912941n
COMMUNICATIONS:
Facile entry to 3d late transition metal boryl complexes
Ba L. Tran, Debashis Adhikari, Hongjun Fan, Maren Pink and Daniel J. Mindiola,
Dalton Trans., 2010, DOI: 10.1039/b912040h
Probing the kinetics of ligand exchange on colloidal gold nanoparticles by surface-enhanced raman
scattering
Yuhua Feng, Shuangxi Xing, Jun Xu, Hong Wang, Jun Wei Lim and Hongyu Chen,
Dalton Trans., 2010, DOI: 10.1039/b912317b
Visit the Dalton Transactions website for more cutting-edge inorganic and organometallic research
www.rsc.org/dalton

PAPER
www.rsc.org/dalton | Dalton Transactions
Three-coordinate late transition metal fluorinated alkoxide complexes†
Stefanie A. Cantalupo,a June S. Lum,a Marisa C. Buzzeo,b Curtis Moore,c Antonio G. DiPasquale,d
Arnold L. Rheingoldc and Linda H. Doerrer*a
Received 1st July 2009, Accepted 17th September 2009
First published as an Advance Article on the web 16th October 2009
DOI: 10.1039/b911758j
Homoleptic fluorinated alkoxide complexes have been prepared from KOC4F9, 1, via salt metathesis
routes. One four-coordinate K{K(18C6)}[Co(OC4F9)4], 2, and four three-coordinate complexes:
{K(18C6)}[Fe(OC4F9)3], 3, {K(18C6)}[Co(OC4F9)3], 4, {K(18C6)}[Cu(OC4F9)3], 5, and
{K(18C6)}[Zn(OC4F9)3], 6, have been prepared and all except 5 have been characterized
crystallographically. Compounds 3, 4, and 6 are very rare examples of monomeric, trigonal alkoxide
complexes. All compounds have been characterized with UV-vis and IR spectroscopy, solution
magnetic susceptibility, and elemental analysis. In solution, compound 2 exists in an equilibrium with 4
and 1, which has been probed with cyclic voltammetry, supporting energetically different Co2+/Co3+
potentials in the three-coordinate (Ep,a = ~1.2 V vs Fc/Fc+) and four-coordinate (Ep,a = ~0.9 V)
geometries. The ligand field engendered by the perfluoro-t-butoxide ligand has been studied with DFT
calculations on 4 and the hypothetical [Co(OC4H9)3]- as well as the previously reported [Co(mes)3]- and
[Co{N(TMS)2}3]- showing significant p-type interactions in the xy plane as well as above and below for
the two alkoxide species.
Fluorinated oxygen-donor ligands are good candidates for stabi-
Introduction
lizing high oxidation states because oxygen atoms are commonly
First row transition metal complexes when highly oxidized have
seen with transition metal complexes with the highest oxidation
-
states, e.g. [Cr2O72-] and [MnO4 ], and extensive fluorination will
a central role in the catalysis of C-H activation followed by se-
lective bond formation, leading to hydrocarbon functionalization.
pull additional electron density away from the metal center. Earlier
n+
=
Voluminous work has been done in pursuit of reactive {M O}
work led to the first complete characterization of homoleptic
fluorinated aryloxide anions of the form [M(OAr)4]2- with Co2+,6
Ni2+,4 and Cu2+.5,6 Aryloxide ligands are potentially susceptible
to oxidation to labile phenoxyl radicals, however, thus attention
shifted to fluorinated alkoxide ligands.
species via O2 activation for alkane functionalization,1 and more
n+
=
recently the isoelectronic imido group {M NR} has also been
investigated. For example the Fe-imido species, [PhBP3]Fe(N-p-
tolyl) can release an imide to CO substrate2 and the Co(III)-
imido species, {TptBu,MeCo NR} demonstrates C-H activation.
3
=
Previous work in homoleptic fluorinated alkoxide metal com-
plexes has been focused on the development of metal oxide pre-
Undesired ligand C-H bond reactivity is a potential hazard in
these complexes. We have chosen to investigate late transition
metal complexes with fluorinated oxygen-donor ligands4-6 whose
C-F bonds are much more robust to oxidation than C-H bonds.
7
cursors for MOCVD, see for example Ba[Cu(OCCH3(CF3)2)3]2
and Na2[Cu(OCH(CF3)2)4].8 We have chosen the perfluoro-t-
butoxide ligand, OC4F9, to remove any ligand C-H bonds from the
immediate coordination environment. To date, this ligand is only
sparsely reported in the literature. A search of the Cambridge
Structural Database (CSD)9 revealed eleven examples of this
ligand bound to a transition metal including V,10 Cu,11,12 Mo,13-16
and Ag17,18 complexes.
aBoston University, Chemistry Department, 590 Commonwealth Ave,
Boston, MA 02215, USA. E-mail: doerrer@bu.edu; Fax: +1 (0)617-353-
6466; Tel: +1 (0)617-358-4335
bNSF Center for Electron Transport in Molecular Nanostructures, Columbia
University, New York, NY 10027, USA. E-mail: mb648@columbia.edu;
Fax: +1 (0) 212-854-1909; Tel: +1 (0) 212-854-6163
The extension of our work to fluorinated alkoxide compounds
began with the preparation of the four-coordinate [Ni(OC4F9)4]2-
anion.4 This work demonstrated that the fluorinated t-butoxide
ligand is the electronic equivalent of fluoride, and therefore a
medium-field ligand, with the potential to stabilize high oxi-
dation states. Here we report further extension of this motif
to cobalt as well as the synthesis of several three-coordinate
compounds with the same ligand. There are numerous examples
in the literature of three-coordinate transition metal complexes,19
however few contain exclusively O-donating ligands and fewer
do so in a monomeric environment. The (tBu3SiO) ligand can
stabilize three coordinate species,20-22 and the [(Ar/RO)M(m2-
OAr/R)2M(OAr/R)] motif is observed with aryloxides with bulky
ortho substituents23-26 or the Ph3CO- anion.27 The majority of
cUniversity of California, San Diego, 9500 Gilman Drive, La Jolla, CA
92093-0358, USA. E-mail: arnrhein@chem.ucsd.edu; Fax: +1 (0)858-
3872; Tel: +1 (0)858-822-3870
dUniversity of California, Berkeley, 32 Lewis Hall, Berkeley, CA 94720-
1460, USA. E-mail: adipasqu@berkeley.edu; Fax: +1 (0)510-642-9675;
Tel: +1 (0)510-642-8444
† Electronic supplementary information (ESI) available: Complete CIFs
for compounds 1–4, 4·THF, and 6·THF as well as ORTEPs for compounds
1 and 4 (Figs. S1 and S2), DFT frequency calculation of 4 (S3), cyclic
voltammogram of 1 (S4), full MO diagram with a- and b-spin orbital
energies for [CoX3]- calculations (S5), MOs from [Co(mes)3]- (S6), MOs
from [Co{N(TMS)2}3]- (S7), as well as tables of interatomic distances
and angles for 1 (Table S1), comparison of X-ray versus DFT determined
distances and angles (S2), and a- and b-spin orbital energies (S3). CCDC
reference numbers 738719–738724. For ESI and crystallographic data in
CIF or other electronic format see DOI: 10.1039/b911758j
374 | Dalton Trans., 2010, 39, 374–383
This journal is
The Royal Society of Chemistry 2010
©

three-coordinate compounds reported with hard donor ligands
utilize only N-donor ligands or a mixture of O- and N-donating
ligands.19 In this contribution we present the synthesis and
characterization of five new three- and four-coordinate complexes
of late, first-row transition metals with the fluorinated alkoxide
ligand perfluoro-t-butoxide.
KOC4F9, 1, in THF to form the tris-perfluoro-t-butoxide metal
complexes. Removal of KX by filtration followed by the addition
of 18-crown-6 in CH2Cl2 resulted in the formation of compounds
3-6 in good yield. Upon initial crystallization, compounds 4 and
6 formed a solvent adduct with a THF molecule in the axial
position, and the X-ray crystal structure of the former has been
obtained. The THF was removed by drying in vacuo and with
further recrystallization in CH2Cl2 and hexanes. The synthesis of
K{K(18C6)}[Co(OC4F9)4], compound 2, was achieved using an
analogous synthesis to the Ni derivative published previously.4
Similar to the Ni case, regardless of the number of equivalents of
crown ether added, only one equivalent of 18-crown-6 is present
in the transition metal product. The second equivalent of K+
is complexed with F atoms from the CF3 groups on the anion
imposing a pseudo three-fold geometry about the K ◊ ◊ ◊ Co axis.
This coordination motif has been seen with similar compounds
and will be discussed in further detail below.
Results and discussion
Syntheses
The synthesis of the ligand potassium salt, KOC4F9, 1, has
been previously reported in the literature,4 however, we present
a procedural improvement. To move away from K0 which could
reductively eliminate KF in the presence of HOC4F9, KH was
used instead to deprotonate the fluorinated alcohol and form the
potassium salt. Colorless crystals are formed in very good yield
when recrystallized from Et2O and hexanes.
The transition metal complexes discussed in this report were
all prepared via a simple metathesis route as shown in Scheme 1.
The metal halides, MX2, are reacted with three equivalents of
Structural characterization
Crystallographic data collection parameters are summarized
in Table 1 and Tables 2 and 3 contain selected interatomic
distances and angles. Compound 1 forms colorless, block-like
crystals and crystallizes as a cubane structure with each K+ atom
bound to three O atoms forming the edges of the cubane as
shown in Fig. S1.† The K+ atoms also interact with F atoms
on the CF3 groups which has been observed previously4 in
K{K(18C6)}[Ni(OC4F9)4] and will be discussed in more detail
below with respect to 2. The average K–O bond distance i◦n 1 is
˚
2.706(2) A, and the average O–K–O bond angle is 81.9(9) (see
Table S1 for more metrical parameters). This cubane motif has
28
=
been seen in other potassium salts such as K{OC( CH2)(mes)},
K(2-OAriPr),29 K(OC(Ph)2Me),29 and tetrakis-(m3-1,3-dineopen-
tylnaphthimidazolin-2-oxo)-bis(THF)-tetra-potassium.30
The
Scheme 1 General metathesis scheme for three- and four-coordinate
metal-alkoxide complex syntheses.
˚
average K–O bond distance in these structures is 2.65(2) A,
Table 1 Summary of X-ray crystallographic data collection parameters
Compound #
1
2
3
4
4·THF
6·THF
Formula
C53H24F72K8O8
2469.52
Monoclinic
P21/c
21.061(11)
21.046(11)
21.082(10)
90
113.734(6)
90
8554(7)
4, 1
1.918
0.608
150(2)
6.79
10.02
100958
15543
C28H24CoF36K2O10
1341.60
Monoclinic
Pn
15.4345(8)
15.2808(8)
19.7039(10)
90
102.4080(10)
90
4538.6(4)
4, 2
1.963
0.760
150(2)
4.71
9.21
68150
13594
C24H24F27FeKO9
1064.38
Triclinic
¯
P1
C24H24CoF27KO9
1067.46
Triclinic
¯
P1
C28H32CoF27KO10
1139.57
Monoclinic
Pc
14.5960(6)
17.7566(8)
17.6882(8)
90
113.9120(10)
90
4190.9(3)
4, 2
1.806
0.679
123(2)
5.02
12.68
34231
11098
C28H32F27KO10Zn
1146.01
Triclinic
¯
P1
Formula Weight
Crystal System
Space Group
˚
a, A
10.4662(16)
10.6115(16)
17.205(3)
82.527(2)
89.910(2)
79.954(2)
1865.1(5)
2, 1
1.895
0.699
100(2)
4.39
10.4471(19)
10.6184(19)
17.165(3)
82.451(3)
89.384(3)
79.214(3)
1854.1(6)
2, 1
1.912
0.759
100(2)
5.78
10.797(3)
10.801(3)
21.012(5)
86.243(3)
84.020(3)
61.203(3)
2135.3(10)
2, 1
1.782
0.840
208(2)
8.07
˚
b, A
˚
c, A
a, deg
b, deg
g , deg
3
˚
V, A
Z, Z¢
r(calc), g cm-3
m(MoKa), mm-1
Temp, K
R(F),%a
R(wF2),%b
Collected
Unique
9.91
38872
6818
11.90
21938
8264
20.87
20819
9460
R(int)
Observed
0.0570
11046
0.0367
12264
0.0368
6322
0.0228
6801
0.0229
10174
0.0375
7580
ꢀ
ꢀ
ꢀ
ꢀ
2
s
2
a R = ꢀFo| - |Fcꢀ/ |Fo|. b R(wF2)={ [w(Fo - Fc2)2]/ [w(Fo2)2]}1/2; w = 1/[s (Fo2) + (aP)2 + bP], P = [2Fc + max(Fo,0)]/3.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 374–383 | 375
©

Table 2 Selected interatomic distances and angles for 2-4
Angle (◦)
˚
Compound
Distance (A)
2
Co–O1
Co–O2
Co–O3
Co–O4
O1–C1
O2–C5
O3–C9
O4–C13
K1 ◊ ◊ ◊ F11
K1 ◊ ◊ ◊ F20
K1 ◊ ◊ ◊ F28
1.875(4)
1.975(3)
1.984(3)
1.974(3)
1.317(6)
1.353(5)
1.345(5)
1.343(5)
2.830(3)
2.855(3)
2.845(3)
O1–Co–O2
O1–Co–O3
O1–Co–O4
O2–Co–O3
O2–Co–O4
O3–Co–O4
Co–O1–C1
Co–O2–C5
Co–O3–C9
Co–O4–C13
124.69(17)
116.59(16)
123.79(16)
92.54(13)
96.32(13)
95.30(13)
167.0(4)
126.1(3)
135.2(3)
134.9(3)
3
Fe–O1
Fe–O2
Fe–O3
O1–C1
O2–C5
O3–C9
Fe ◊ ◊ ◊ O8
K ◊ ◊ ◊ O1
K ◊ ◊ ◊ O2
1.9595(17)
1.9782(17)
1.8748(19)
1.347(3)
1.352(3)
1.343(3)
2.2260(17)
2.9992(19)
2.9706(18)
O1–Fe–O2
O1–Fe–O3
O2–Fe–O3
Fe–O1–C1
Fe–O2–C5
Fe–O3–C9
100.90(7)
136.41(9)
121.83(9)
143.35(16)
134.65(15)
154.54(19)
Fig. 1 ORTEP diagram of [Co(OC4F9)4]2-, 2, with K+ cation interaction
with F atoms. K+ complexed with 18-crown-6 removed for clarity.
Ellipsoids are shown at the 50% probability level.
4
Co–O1
Co–O2
Co–O3
Co–O4
O1–C1
O2–C5
O3–C9
K1 ◊ ◊ ◊ O1
K1 ◊ ◊ ◊ O3
1.9143(18)
1.8810(19)
1.9262(18)
2.2222(17)
1.343(3)
1.342(3)
1.355(3)
O1–Co–O2
O1–Co–O3
O2–Co–O3
Co–O1–C1
Co–O2–C5
Co–O3–C9
128.38(9)
107.23(8)
124.39(9)
145.74(17)
150.27(18)
133.66(16)
˚
an average distance of 2.944(3) A, similar to the Ni analogue
˚
of this complex which contains 2.709(2) A contacts in the same
arrangement.4 The K–F interactions result in an approximate
three-fold geometry in both Co and Ni cases. Aside from the
analogous Ni complex, a search of the CSD produced 17 structures
that exhibit the same K–F interaction to CF3 groups in the range
3.115(2)
3.0056(19)
˚
of 2.7-3.5 A. In total, 73 K–F interactions were compared and the
˚
average distance was 2.9(2) A. The values for the K–F contacts for
compound 2 fall within this range.
Crystallographic characterization of compounds 3 and 4,
reveals both anions to have three t-butoxide ligands bound in
a trigonal planar geometry around the metal center. In compound
Table 3 Selected interatomic distances and angles for 4·THF and 6·THF
˚
Compound
Distance (A)
Angle (◦)
˚
3, shown in Fig. 2, the average Fe–O bond distance is 1.94(6) A
with O–Fe–O bond angles ranging from 100.90(7)◦ to 136.41(9)◦
4·THF
Co–O7
Co–O8
Co–O9
Co–O10
O7–C13
O8–C17
O9–C21
1.889(4)
1.928(3)
1.923(3)
2.065(4)
1.321(6)
1.334(6)
1.328(6)
O10–Co–O7
O10–Co–O8
O10–Co–O9
O7–Co–O8
O7–Co–O9
O8–Co–O9
108.11(19)
90.73(15)
92.52(15)
114.4(2)
121.3(2)
119.74(16)
and the Fe center lies within the plane of the O atoms. Compound
˚
4 (Fig. S2†) has an average Co–O bond distance of 1.91(2) A with
O–Co–O bond angles ranging from 107.23(8)◦ to 128.38(9)◦ and
the Co center lies directly in the best plane of the O atoms. Both
anions have elongated interactions with the K+ atom complexed by
18-crown-6 to two of the oxygen atoms bound to the metal centers,
for example in compound 3 the K ◊ ◊ ◊ O1 distance is 2.9992(19) and
K ◊ ◊ ◊ O2 distance is 2.9706(18) as seen in Fig. 2 (top).
6·THF
Zn–O1
Zn–O2
Zn–O3
Zn–O4
O1–C1
O2–C5
O3–C9
1.910(3)
1.892(3)
1.896(3)
2.091(3)
1.345(4)
1.337(4)
1.330(5)
O4–Zn–O1
O4–Zn–O2
O4–Zn–O3
O1–Zn–O2
O1–Zn–O3
O2–Zn–O3
98.00(12)
94.95(12)
98.90(12)
120.44(12)
115.84(12)
119.01(13)
Compounds 4·THF and 6·THF have approximately trigonal
pyramidal coordination with three basal alkoxide ligands and an
axial THF molecule, as shown in Fig. 3. Compound 4·THF has
˚
a Co–OTHF distance, 2.070(3) A which is considerably longer than
˚
the average Co–Oalkoxide distance of 1.925(3) A. As seen in Table 3,
suggesting a slightly longer K–OR bond in the fluorinated case
due to the additional electron-withdrawing substituents.
the basal plane O–Co–O bond angles between the alkoxide ligands
are similar with an average of 118.03(14)◦. The average OTHF–Co–
Oalkoxide angle is 98.08(14)◦ such that the trigonal planar geometry
of the alkoxide ligands around the metal center is largely conserved
The transition metal complexes described in this paper show
three different types of geometries around the metal center based
on differences in ligand stoichiometry and solvent coordination.
The addition of four alkoxide ligands to CoI2 effects a four-
coordinate, pseudo-tetrahedral geometry around the metal center
of compound 2, as seen in Fig. 1, which shows the strong K+
interaction with the F atoms on several CF3 groups. Interestingly,
this K+ atom favors this type of coordination instead of being
complexed by 18-crown-6. These K–F contacts are present at
˚
with the Co atom displaced only 0.237 A out of the best O3 plane.
˚
Compound 6·THF has a Zn–OTHF distance of 2.091(3) A which is
˚
also longer than the average Zn–Oalkoxide distance of 1.899(3) A. The
average O–Zn–O bond angle for the alkoxide ligands is 118.43(12)◦
while the average OTHF–Zn–Oalkoxide angle is 97.28(12)◦. Again the
˚
displacement of the Zn atom from the best O3 plane (0.240 A) is
minimal.
376 | Dalton Trans., 2010, 39, 374–383
This journal is
The Royal Society of Chemistry 2010
©

Table 4 Color and magnetic susceptibility data for selected three-
coordinate complexes
Compound
Color
meff (mB) CSD code
Reference
Fe
[Fe(OC4F9)3]- (3)
[Fe(mes)3]-
off-white
colorless
green
yellow
lt brown
4.23
5.56
4.6
3.4
3.9
—
—
This work
40
[Fe(N(SiMe3)3)3]-
[Fe2(2,4,6-OArtBu)4]
[Fe2(OB(mes)2)4]
GEKKUC
JIXKIK
JIXLOR
41,42
25
43
Co
[Co(OC4F9)3]- (4)
[Co(mes)3]-
green
4.49
3.8
—
—
—
—
—
This work
44,45
lt green
dk green
dk green
green
[Co(N(SiMe3)3)3]-
[Co2(OC(C6H11)3)4]
[Co2(OC(Ph)3)4]
TEGCEN
FIRSII
FIRSOO
42
27
27
Cu
[Cu(OC4F9)3]- (5)
green
1.87
—
—
KIWZIZ
This work
[Cu(OC(CF3)2CH3)3]- green
7
Zn
[Zn(OC4F9)3]- (6)
[Zn2(2,6-OArtBu)4]
[Zn(2,6-OArtBu)3]-
colorless
colorless
colorless
—
—
—
—
This work
46
DORRAD
HIWLAA
47
Electronic spectroscopy
The electronic spectra for the three-coordinate complexes 3-5 are
shown in Fig. 4 and are compared with the reported data on
related literature compounds in Table 4. Compound 3 is off-white
in color with the other three-coordinate Fe2+ complexes ranging
from colorless to light brown. The color of compound 4 is green,
similar to other three-coordinate Co2+ complexes in the literature.
The green color exhibited by a three-coordinate Cu2+ complex with
a related tertiary alkoxide ligand7 is the same as compound 5.
Fig. 2 ORTEP diagrams of compound 3. Top: [Fe(OC4F9)3]- with K+
complexed with 18-crown-6 cation (hydrogen atoms removed for clarity).
Bottom: [Fe(OC4F9)3]- (cation removed for clarity). Ellipsoids are shown
at the 50% probability level.
Fig. 3 ORTEP diagram of 4·THF (left) and 6·THF (right). Cations and
F atoms removed for clarity. Ellipsoids are shown at the 50% probability
level.
Fig. 4 UV-vis spectra of {K(18C6)}[Fe(OC4F9)3], 3, {K(18C6)}[Co-
(OC4F9)3], 4, and {K(18C6)}[Cu(OC4F9)3], 5.
Although crystals suitable for X-ray diffraction experiments on
5 have not been grown, we propose a similar monomeric, trigonal
planar structure for the Cu derivative as well. A mononuclear
species is quite likely based on the structures of 3, 4, and 6,
as well as the previously reported [Cu4(OC4H9)6(OC4F9)2] in
which the fluorinated alkoxides occupy the terminal positions and
the OC4H9 ligands occupy bridging sites in an [(F9C4O)Cu{(m2-
OC4H9)Cu}3(OC4F9)] tetramer.11
Initial dissolution of 2 in CH2Cl2, red trace in Fig. 5, resulted in
a spectrum approximately similar to the spectra for other [CoX4]2-
species, showing three maxima in the range 500–700 nm.6 When
an additional equivalent of ligand is added, however, the resulting
spectrum shifted to that of the blue trace in Fig. 5. This result
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 374–383 | 377
©

[Co(OArF)4]2- complexes which have magnetic moments in the
range of 4.45-5.02 mB.6 The magnetic moment of [Fe(OC4F9)3]-,
3, is 4.23 mB which is smaller than the spin-only moment for four
unpaired electrons, 4.89 mB. The cobalt derivative [Co(OC4F9)3]-,
4, has a magnetic moment of 4.49 mB again larger than the expected
magnetic moment for a spin-only moment for three unpaired
electrons. Lastly the anion [Cu(OC4F9)3]-, 5, exhibited a magnetic
moment of 1.87 mB, similar to the expected S = 12 spin-only moment
of 1.73 mB.
Electrochemistry
Preliminary cyclic voltammetric studies of the two complexes 4 and
2 show definite redox activity of the cobalt centers, as shown in
Figs. 6–8. Furthermore, it is evident that 4 and 2 are in equilibrium
on the voltammetric time scale, supporting ligand dissociation
from 2 to form 4 as posited based on the UV-vis spectra.
Comparison of the voltammogram of 4 in Fig. 6 with that of 2
in Fig. 7 shows that for both complexes the same features, labeled
with roman numerals, are observed in dramatically different ratios:
Fig. 5 UV-vis spectra of K{K(18C6)}[Co(OC4F9)4], 2, (red) and 2 with
an additional equivalent of ligand, (blue).
suggests that compound 2 may undergo ligand dissociation in solu-
tion to form some amount of 4. Crystallographic studies confirmed
that compound 2 is a four-coordinate, pseudo-tetrahedral Co2+
compound which when dissolved in non-coordinating solvents
forms a blue solution distinct from that of green 4. When
additional ligand is present, however, a different spectrum is
observed, suggesting that any ligand dissociation is inhibited and
all Co centers have a four-coordinate geometry in solution. The
dissociation/association of ligands will be discussed further below
in the context of the electrochemical data.
a quasi-reversible couple at significantly high potentials (Emid
~1.2 V, peaks I and II) and peaks appearing at ca. +0.89 V (III)
=
and +0.22 (IV) on the forward and reverse scans, respectively.
Vibrational spectroscopy
The assignment of M–O stretching frequencies is a challenge
in metal-alkoxide complexes due to the coupling of C–O and
M–O modes.31 Few data have been assigned to specific bands for
terminal alkoxide ligands on first row transition metals, however,
Fe(OiPr)3 and Fe(OnBu)3 have been reported to have stretching
frequencies at 603 and 675 cm-1.32 A survey of various homoleptic
metal ethoxide species (e.g. Al(III) and Ta(V)) reported frequencies
for the M–O stretch from 392 to 703 cm-1 (see Table 2.18 in
ref. 33). In the extensively studied anion,34-36 [Al(OC4F9)4]-, the
vibrational data37 show characteristic absorptions at 727 cm-1
(C–C, C–O) and 974 cm-1 (C–C, C–F) along with other bands
in the range from 1100 to 1400 cm-1.37 The complexes discussed
in this paper all contain the characteristic t-butoxide stretching
frequencies around 724 and 962 cm-1 as well as other bands from
1100 to 1400 cm-1. The calculated frequencies for compound 4
are shown in the Supplementary Information (Figure S3†) and
include bands centered at 1210, 978, and 718 cm-1, consistent with
the experimental data. Due to the limitations of the instrument,
vibrational experimental studies only extend to 500 cm-1.
Fig. 6 Cyclic voltammogram of 5 mM {K(18C6)}[Co(OC4F9)3], 4, in
0.1 M TBAPF6/CH2Cl2.
Magnetism
The Evans method38,39 was used to collect magnetic solution
susceptibility measurements which are listed in Table 4. The
magnetic susceptibility measurement for [Co(OC4F9)4]2-, 2, is
larger than expected for the spin-only moment of 3.87 mB, with
a value of 4.66 mB, but is similar to other compounds previously
reported that have a similar Co–O4 coordination, particularly the
Fig. 7 Cyclic voltammogram of K{K(18C6)}[Co(OC4F9)4], 2, in 0.1 M
TBAPF6/CH2Cl2.
The large peak splitting observed between III and IV suggests
the complex in question undergoes either geometric and/or
378 | Dalton Trans., 2010, 39, 374–383
This journal is
The Royal Society of Chemistry 2010
©

the window of complex activity (see Fig. S4†). Further studies are
focused on deciphering the mechanistic details of the individual
redox processes.
Electronic structure calculations
Computational studies have been undertaken to inform our
understanding of the {MO3} coordination environment and the
effect of ligand fluorination on the metal–ligand interaction. In
particular, compound 4 with enforced C3h symmetry was studied in
comparison to the known [Co(mes)3]-,45 [Co{N(TMS)2}3]-,42 and
the hypothetical [Co(OC4H9)3]-. For all four species, C3 symmetry
was imposed initially and the two alkoxide derivatives also had
mirror plane symmetry. The rotational symmetry relaxed in the
alkoxide and amide calculations, as can be seen in Table S2,†
because the ADF package does not support calculations in point
groups with irrational number characters. The mirror plane is not
observed in the X-ray diffraction structure of 4 because of the
Fig. 8 Cyclic voltammograms of K(18C6)}[Co(OC4F9)3], 4, alone (black
trace) and in the presence of an additional equivalent of ligand, (red trace)
in 0.1 M TBAPF6/CH2Cl2.
+
interaction with the {K(18C6)} cation, but was imposed to keep
electronic rearrangement upon oxidation. Rearrangement be-
tween three-coordinate trigonal planar and four-coordinate tetra-
hedral geometries is such a change. In the case of 4, the quasi-
reversible couple (I/II) appears almost exclusively, while peaks III
and IV dominate the voltammetric profile of 2. Accordingly, peak
I is tentatively ascribed to the oxidation of 4 and peak III to the
oxidation 2.
the four systems as similar as possible. A compound with {MX3}
coordination and strong s-donor ligands in the xy plane but no
appreciable p-interactions would result in a ligand-field splitting
as shown on the left of Scheme 3. When there are M-L p-type
interactions above and below the xy plane, then the dxz and dyz
orbitals are expected to be destabilized to some extent as shown
on the right of Scheme 3. The results of our DFT study are shown
in Fig. 9 in which the a- and b-spin orbital energies (eV) are
averaged. All individual spin orbital energies are shown in Fig. S5
in the supplementary information and tabulated in Table S3.† The
[Co(mes)3]- species is leftmost in Fig. 9 and shows an important
difference versus the s-only picture in Scheme 3. The (dxz, dyz) pair
are raised in energy above the dz2 orbital because of an overlap
with the p* system of the mesityl ring as shown in supplementary
information Fig. S6.† As expected, the [Co{N(TMS)2}3]- anion
also shows a destabilized (dxz, dyz) pair, this time from interaction
with the N lone pair, Fig. S7.† The two [Co(OR)3]- splitting
diagrams are further differentiated from the first two cases. These
differences arise because there are now two lone pairs on each
donor atom and they interact with both the (dxz, dyz) orbitals and
These results suggest that a small amount of 2 is present in
a solution of 4. Given our confidence in the purity of 4 (triply
recrystallized), in conjunction with the spectroscopic data, it
is most plausible that small quantities of the four-coordinate
complex are generated upon dissolution of 4 in CH2Cl2 with nearly
spontaneous ligand dissociation and reassociation (Scheme 2). To
further support this claim, complex 2 was then generated in situ by
addition of excess ligand 1 to 4 and examined voltammetrically.
As can be seen in Fig. 8, upon introduction of 1, the peaks at
less anodic potentials grow in considerably, with concomitant
diminution of the quasi-reversible couple I/II. We assign the
activity observed in features I and II (Emid = ~1.2 V) and features
III and IV (Ep,a = ~ +0.9 V; Ep,c = ~ +0.2 V) to the Co2+/Co3+
couple in the three- and four-coordinate complexes respectively.
The formation of 2 from 4 implies that a {Co-OC4F9} species
with a Co:OR ratio of 1:2, or its stoichiometric equivalents, must
be present. No such species has been isolated to date, however
the existence of an [Co2X4] derivative such as [(F9C4O)Co(m2-
OC4F9)2Co(OC4F9)] has precedent in the structure of the anal-
ogous amide [{(TMS)2N}Co(m2-{N(TMS)2}2Co{N(TMS)2}],48
and the [(Ar/RO)M(m2-OAr/R)2M(OAr/R)] species. Interest-
ingly, complex 4 is much harder to oxidize than 2, suggesting
that either a fourth ligand or a different coordination geometry are
required to stabilize the Co3+ species. Control studies on the ligand,
KOC4F9, 1, show an irreversible small oxidation wave outside of
2
2
the (dxy, dx ) orbitals. These overlaps are shown for the a-spin
-y
orbitals of [Co(OC4H9)3]- in Fig. 10.
Scheme 3 Qualitative ligand-field splitting for d-manifold in three-fold
symmetry.
Our previous work4-6 has demonstrated that extensive fluorina-
tion of aryloxide ligands made possible monodentate phenolate
ligands in the absence of bulky ortho substituents. We proposed
that the fluorination caused sufficient reduction in the ligand
Scheme 2 Ligand association/dissociation equilibria.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 374–383 | 379
©

Fig. 9 Comparison of d-orbital manifold energies for [CoX3]-.
Experimental
General information
Reactions and NMR sample preparations were performed at room
temperature in a glovebox or on a Schlenk line under an atmo-
sphere of purified N2. The anhydrous solvents tetrahydrofuran
(THF), dichloromethane (CH2Cl2), diethyl ether (Et2O), toluene,
and hexanes were purified using an argon-filled MBraun solvent
purification system equipped with columns of anhydrous Al2O3.
Acetonitrile (MeCN) and fluorobenzene (FPh) were dried by
refluxing over CaH2 and distilled under N2. 1,2-Dimethoxyethane
(DME) was dried using sodium-benzophenone ketyl and dis-
tilled under N2. Deuterated solvents used for NMR samples
(CDCl3 and CD2Cl2) and the internal 19F reference standard,
trichlorofluoromethane (CFCl3) were dried by refluxing over CaH2
and stored in an N2-filled glovebox. Molecular sieves were used
to dry d6-acetone which was stored in an N2-filled glovebox.
All solvents were stored over molecular sieves in the glovebox.
Celite was dried overnight in vacuo while being heated to 125 ◦C
with an oil bath. 18-crown-6 was obtained commercially and
recrystallized from toluene and hexanes. All other reagents were
obtained commercially and used without any further purification.
NMR spectra were measured on a Varian 300 MHz spectrometer.
1H and 13C chemical shifts were referenced to the resonance of
residual protiosolvent (1H) or the 13C resonance of the solvent.
19F shifts were referenced to internal CFCl3. Solution phase
magnetic susceptibilities were determined via the Evans method
in d6-acetone or CD2Cl2 with 10% (Me3Si)2O and the same
solution as an internal reference and reported after appropriate
diamagnetic corrections.51,52 UV-vis data were collected with a
Shimadzu UV-3600 spectrometer. Vibrational spectroscopy data
Fig. 10 View down z-axis on d-orbitals with M-L p interactions in
[Co(OC4H9)3]-.
p-donor character that the likelihood of aryloxide-metal center
bridging was no longer certain. This reduction in p-donor
character is evident in the comparison of the two rightmost
splitting diagrams in Fig. 9. The [Co(OC4F9)3]- orbitals show less
increase in energy of the (dxz, dyz) orbitals versus the [Co(OC4H9)3]-
species. Notably, the only structurally characterized transition
metal compounds with an {M(OC4H9)3} unit show extensive
bridging by the alkoxide.49,50
380 | Dalton Trans., 2010, 39, 374–383
This journal is
The Royal Society of Chemistry 2010
©

was collected on a Nicolet Nexus 670 FT-IR with an attenuated
total reflectance accessory. Microanalyses were performed by H.
Kolbe Microanalytisches Laboratorium, Mu¨lheim an der Ruhr,
Germany and Quantitative Technologies, Inc., Lebanon, New
Jersey.
overnight. The solution was then filtered through Celite to remove
any residual KI, concentrated to dryness, and the product was
recrystallized from CH2Cl2 and hexanes. Emerald green crystals
were isolated in 48% yield. A THF adduct crystallized as a blue
crystal. THF was removed by drying in vacuo and recrystallization.
UV-vis (CH2Cl2) (lmax, nm (e, cm-1 M-1)) 660 (122), 467 (19). IR
(cm-1): (neat) 1478, 1354, 1263, 1234, 1195, 1109, 963, 837, 725,
668. Anal. Calcd for KC24H24O9F27Co: C, 27.01; H, 2.27; F, 48.06.
Found: C, 27.00; H, 2.07; F, 48.21. meff (Evans’ method, CD2Cl2) =
4.49 mB.
Synthetic procedures
KOC4F9, 1. Synthesis and characterization data have been
previously reported,4 but we provide here an improved method. A
portion of KH (1.876 g, 46.76 mmol) was slurried in anhydrous
Et2O and cooled to -78 ◦C under N2. HOC4F9 (10.158 g,
43.04 mmol) was added via syringe while the reaction mixture
was stirring. The reaction mixture turned from clear and colorless
to very pale yellow and cloudy. The reaction mixture was left to
warm to room temperature and stir for three hours forming a
cloudy off-white solution. The Et2O was removed in vacuo, the
flask was brought into the drybox and the crude product was
redissolved in Et2O and filtered through Celite. The product was
recrystallized from Et2O and hexanes and colorless crystals were
collected in 99% yield (11.538 g).
{K(18C6)}[Cu(OC4F9)3], 5. A solution of KOC4F9 (1.003 g,
3.66 mmol) was added to an orange-brown slurry of CuBr2
(0.276 g, 1.23 mmol) in THF. The reaction mixture turned
cloudy yellow-green and was left stirring at room temperature
overnight. The cloudy green solution was filtered through Celite
to remove KBr and triturated three times with CH2Cl2. A
solution of 18-crown-6 (0.322 g, 1.22 mmol) was added in
CH2Cl2 and the resultant green solution was stirred overnight
at room temperature. The solution was filtered to remove residual
KBr, CH2Cl2 was removed in vacuo and the crude product was
recrystallized from CH2Cl2 and hexanes. The recrystallized green
product was isolated in 74.7% yield (0.976 g). UV-vis (CH2Cl2)
(lmax, nm (e, cm-1 M-1)): 305 (4030), 845 (335). IR (cm-1): (neat)
1354, 1264, 1231, 1181, 1104, 962, 837, 724. Anal. Calcd for
{{K(18C6)}[Cu(OC4F9)3]+0.5CH2Cl2}; KC24.5H25O9ClF27Cu: C,
26.40; H, 2.26; F, 46.03. Found: C, 26.39; H, 2.01; F, 46.28. meff
(Evans’ method, CD2Cl2) = 1.87 mB.
K{K(18C6)}[Co(OC4F9)4], 2. A portion of CoI2 (143 mg,
0.457 mmol) was combined with KOC4F9 (501 mg, 1.826 mmol)
in 10 mL THF following the same procedure as for
K{K(18C6)}[Ni(OC4F9)4],4 including the addition of 1 equiv of
18-crown-6. The crude product was recrystallized from CH2Cl2
and hexanes and bright blue crystals were collected in 65% yield.
UV-vis (CH2Cl2) (lmax, nm (e, cm-1 M-1)) 634 (77), 581 (68), 526
(43). IR (cm-1): (neat) 1354, 1227, 1191, 1106, 959, 837, 723, 668.
Anal. Calcd for K2C28H24O10F36Co: C, 25.07; H, 1.80; F, 50.98.
Found: C, 24.90; H, 1.71; F, 50.95. meff (Evans’ method, CD2Cl2) =
4.66 mB
{K(18C6)}[Zn(OC4F9)3], 6. A portion of ZnI2 (196 mg,
0.614 mmol) was combined with KOC4F9 (503 mg, 1.836 mmol)
in ~10 mL of THF resulting in the formation of a white precipitate.
The cloudy white reaction mixture was left to stir overnight at
room temperature after which time was filtered through Celite
to remove KI. The clear and colorless filtrate was concentrated
under vacuum to an off-white solid and triturated three times
with CH2Cl2. An equivalent of 18-crown-6 (162 mg, 0.611 mmol)
was added in CH2Cl2 and the yellow reaction mixture was
left to stir overnight at room temperature. The solution was
filtered through Celite to remove residual KI and the filtrate
was dried to an off-white/yellowish solid. The product was
recrystallized from CH2Cl2 and hexanes and colorless crystals
were collected in 85% yield. IR (cm-1): (neat) 1352, 1247, 1194,
{K(18C6)}[Fe(OC4F9)3], 3. A portion of FeI2 (566 mg,
1.827 mmol) and KOC4F9 (1.503 g, 5.481 mmol) were combined
and dissolved in THF which resulted in the immediate formation
of a cloudy bright yellow reaction mixture that was left to stir
at room temperature overnight. The yellow slurry was filtered
through Celite to remove KI, the filtrate was concentrated to
dryness under vacuum, and the product was triturated three times
with FPh. An equivalent of 18-crown-6 (481 mg, 1.820 mmol)
was added and the yellow solution was left to stir at room
temperature overnight. The solution was filtered through Celite
to remove residual KI and concentrated under vacuum to an
off-white powder. The product was recrystallized from FPh and
hexanes and off-white colorless crystals were collected in 75%
yield. UV-vis (Et2O) (lmax, nm (e, cm-1 M-1)) 248 (436). IR (cm-1):
(neat) 1355, 1265, 1234, 1191, 1108, 963, 725, 668. Anal. Calcd for
KC24H24O9F27Fe: C, 27.08; H, 2.27; F, 48.19. Found: C, 26.68; H,
2.32; F, 48.16. meff (Evans’ method, d6-acetone) = 4.23 mB.
1
1105, 961, 838, 724. H NMR (d, ppm, {d6-acetone}): 3.65 (s,
1
24H, CH2). 13C{ H} NMR (d, ppm, {d6-acetone}): 70.86 (s,
CH2). 19F NMR (d, ppm, {d6-acetone}): 75.18 (s). Anal. Calcd
for {{K(18C6)}[Zn(OC4F9)3]·1.0CH2Cl2};KC25H26O9Cl2F27Zn:
C, 25.91; H, 2.26; F, 44.27. Found: C, 25.81; H, 2.20; F, 44.32.
X-ray crystallography
{K(18C6)}[Co(OC4F9)3], 4. A portion of CoI2 (190.1 mg,
0.6078 mmol) was combined with KOC4F9 (499 mg, 1.820 mmol)
in 10 mL of THF and the formation of a blue-green mixture and
white precipitate was observed. The reaction mixture was stirred
at room temperature overnight after which time the solution was
filtered through Celite to remove KI. The filtrate was concentrated
to dryness and triturated three times with CH2Cl2. An equivalent
of 18-crown-6 (161 mg, 0.608 mmol) was added in CH2Cl2 and
the blue-green solution was left to stir at room temperature
A summary of crystal data collection and refinement parameters
for all compounds are given in Table 1. Selected interatomic
distance and angle parameters are given in Tables 2 and 3. All
data were collected on APEX-CCD-detector equipped Bruker
diffractometers, and were corrected for absorption using semi-
empirical, multi-scan methods. All structures were solved by
heavy-atom methods and the remaining non-hydrogen atoms were
located from subsequent difference maps. All structures were
refined with anisotropic thermal parameters for all non-hydrogen
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 374–383 | 381
©

atoms; hydrogen atoms were treated as idealized contributions. All
software is contained in various libraries (SHELXTL, SMART
and SAINT) maintained by Bruker AXS, Madison, WI.53 Com-
pound 2 is a racemic twin and has been so refined.
ligands may facilitate substitution by more strongly bound,
multiply bonding groups (e.g. terminal oxo or imido functional
groups).
Acknowledgements
Electrochemistry
This work was supported by Boston University, the Petroleum
Research Foundation (ACS PRF 48022-AC3) and the National
Science Foundation (purchase of NMR spectrometer CHE
0619339). MCB thanks the Nanoscale Science and Engineering
Initiative of the National Science Foundation under NSF Award
Number CHE-0641523, and the New York State Office of Science,
Technology, and Academic Research (NYSTAR).
Electrochemical experiments were performed with 0.1 M TBAPF6
(triply recrystallized from CH2Cl2 and hexanes) in CH2Cl2 with
5.0 mM analyte concentrations. All experiments were performed
in an N2-filled drybox, using a three-electrode cell connected
through binding posts to an external CHI 630C potentiostat run
by a personal computer with CHI software. All voltammograms
shown were recorded at a scan rate of 100 mV s-1. A glassy carbon
macroelectrode (2 mm diameter) was employed as the working
electrode, with Ag and Pt wires as quasi-reference and counter
electrodes, respectively. Prior to each experiment, the working
electrode was polished using 0.05 mm polishing alumina on a
microcloth polishing pad and rinsed with CH2Cl2. All potentials
are reported versus Fc/Fc+.
References
1 Activation of Small Molecules, ed. W. B. Tolman, Wiley-VCH, Wein-
heim, 2006, pp. 363.
2 S. D. Brown, T. A. Betley and J. C. Peters, J. Am. Chem. Soc., 2003,
125, 322–323.
3 D. T. Shay, G. P. A. Yap, L. N. Zakharov, A. L. Rheingold and K. H.
Theopold, Angew. Chem., Int. Ed., 2005, 44, 1508–1510.
4 B. N. Zheng, M. O. Miranda, A. G. DiPasquale, J. A. Golen, A. L.
Rheingold and L. H. Doerrer, Inorg. Chem., 2009, 48, 4274–4276.
5 M. V. Childress, D. Millar, T. M. Alam, K. A. Kreisel, G. P. A. Yap,
L. N. Zakharov, J. A. Golen, A. L. Rheingold and L. H. Doerrer, Inorg.
Chem., 2006, 45, 3864–3877.
6 M. C. Buzzeo, A. H. Iqbal, C. M. Long, D. Millar, S. Patel, M. A.
Pellow, S. A. Saddoughi, A. L. Smenton, J. F. C. Turner, J. D.
Wadhawan, R. G. Compton, J. A. Golen, A. L. Rheingold and L. H.
Doerrer, Inorg. Chem., 2004, 43, 7709–7725.
7 A. P. Purdy and C. F. George, Inorg. Chem., 1991, 30, 1969–1970.
8 A. P. Purdy, C. F. George and J. H. Callahan, Inorg. Chem., 1991, 30,
2812–2819.
9 F. H. Allen, Acta Crystallogr., Sect. B: Struct. Sci., 2002, 58, 380–388.
10 V. J. Murphy and H. Turner, Organometallics, 1997, 16, 2495–2497.
11 A. P. Purdy, C. F. George and G. A. Brewer, Inorg. Chem., 1992, 31,
2633–2638.
Electronic structure calculations
DFT calculations used a QS8-2400C computer from Parallel
Quantum Solutions (PQS) with the Amsterdam Density Func-
tional Package 2008.0154-56 on all electron, geometry optimized
systems with TZ2P basis sets and the Vosko, Wilk, and Nusair57
local density approximation. Calculations were carried out on
four species with symmetry enforced as indicated: [Co(OC4F9)3]-
C3h (Cs), [Co(OC4H9)3]- C3h (Cs), [Co(mes)3]- C3, (no sym) and
[Co{N(TMS2)}3]- C3 (no sym). A comparison of intramolecular
distances and angles from crystallographic studies (except for
the hypothetical [Co(OC4H9)3]-) with the computational ones is
presented in Table S3.† An S = 3/2 spin state was used in an
unrestricted spin calculation in each case.
12 C. George and A. Purdy, Acta Crystallogr., Sect. C: Cryst. Struct.
Commun., 1992, 48, 155–157.
13 R. L. Gdula, M. J. A. Johnson and N. W. Ockwig, Inorg. Chem., 2005,
44, 9140–9142.
Conclusions
14 R. G. Abbott, F. A. Cotton and L. R. Falvello, Polyhedron, 1990, 9,
The synthesis of three- and four-coordinate transition metal
fluorinated alkoxide complexes are reported with thorough spec-
troscopic, crystallographic, and electronic structure characteriza-
tion. The fluorinated alkoxide ligand does not bridge transition
metal centers but is a terminal ligand in monomeric species. An
interesting equilibrium between the four-coordinate Co complex,
2, and the three-coordinate Co complex, 4, and an equivalent
of ligand is suggested based on UV-visible spectroscopy data
in conjunction with electrochemical data. Cyclic voltammetric
studies also suggest a notably higher Co2+/Co3+ couple in the case
of 4 versus 2, indicating the fourth alkoxide ligand does indeed
serve to afford stabilization to the oxidized product. Electronic
structure calculations of 4 and related trigonal [CoX3]- species
reveal substantial p-type interactions of the alkoxide oxygen atom
lone pairs, not only above and below the xy plane as seen for an
amide derivative, but also in the xy plane.
1821–1827.
15 D. A. Johnson, J. C. Taylor and A. B. Waugh, J. Inorg. Nucl. Chem.,
1980, 42, 1271–1275.
16 D. A. Johnson, J. C. Taylor and A. B. Waugh, Inorg. Nucl. Chem. Lett.,
1979, 15, 205–206.
17 A. Reisinger, D. Himmel and I. Krossing, Angew. Chem., Int. Ed., 2006,
45, 6997–7000.
18 A. Reisinger, N. Trapp and I. Krossing, Organometallics, 2007, 26,
2096–2105.
19 C. C. Cummins, Prog. Inorg. Chem., 1997, 47, 685–836.
20 P. T. Wolczanski, Chem. Commun., 2009, 740–757.
21 O. L. Sydora, D. S. Kuiper, P. T. Wolczanski, E. B. Lobkovsky, A.
Dinescu and T. R. Cundari, Inorg. Chem., 2006, 45, 2008–2021.
22 O. L. Sydora, P. T. Wolczanski, E. B. Lobkovsky, C. Buda and T. R.
Cundari, Inorg. Chem., 2005, 44, 2606–2618.
23 D. J. Darensbourg, J. R. Wildeson, S. J. Lewis and J. C. Yarbrough,
J. Am. Chem. Soc., 2002, 124, 7075–7083.
24 G. B. Deacon, P. E. Fanwick, A. Gitlits, I. P. Rothwell, B. W. Skelton
and A. H. White, Eur. J. Inorg. Chem., 2001, 1505–1514.
25 R. A. Bartlett, J. J. Ellison, P. P. Power and S. C. Shoner, Inorg. Chem.,
1991, 30, 2888–2894.
26 P. B. Hitchcock, M. F. Lappert and A. Singh, J. Chem. Soc., Chem.
Commun., 1983, 1499–1501.
27 G. A. Sigel, R. A. Bartlett, D. Decker, M. M. Olmstead and P. P. Power,
Inorg. Chem., 1987, 26, 1773–1780.
28 X. He, B. C. Noll, A. Beatty, R. E. Mulvey and K. W. Henderson,
J. Am. Chem. Soc., 2004, 126, 7444–7445.
Reactivity explorations of the low-coordinate complexes re-
ported here are underway, including one-electron oxidations, oxo
and imido group transfer reactions, and possible metal reductions.
The design of the ligand is favorable for the (perhaps transient)
stabilization of highly oxidized metal centers that are immune to
internal C–H activation and the lability of some of the alkoxide
382 | Dalton Trans., 2010, 39, 374–383
This journal is
The Royal Society of Chemistry 2010
©

29 T. J. Boyle, N. L. Andrews, M. A. Rodriguez, C. Campana and T. Yiu,
Inorg. Chem., 2003, 42, 5357–5366.
43 H. Chen, P. P. Power and S. C. Shoner, Inorg. Chem., 1991, 30, 2884–
2888.
44 K. H. Theopold, J. Silvestre, E. K. Byrne and D. S. Richeson,
Organometallics, 1989, 8, 2001–2009.
45 W. Seidel and I. Burger, Z. Chem., 1977, 17, 31.
46 M. Kunert, M. Brauer, O. Klobes, H. Gorls, E. Dinjus and E. Anders,
Eur. J. Inorg. Chem., 2000, 1803–1809.
47 D. J. Darensbourg, S. A. Niezgoda, J. D. Draper and J. H. Reibenspies,
Inorg. Chem., 1999, 38, 1356–1359.
30 S. Saravanakumar, A. I. Oprea, M. K. Kindermann, P. G. Jones and J.
Heinicke, Chem.–Eur. J., 2006, 12, 3143–3154.
31 D. C. Bradley, R. C. Mehrotra and D. P. Gaur, Metal Alkoxides,
Academic Press, New York, 1978, pp. 411.
32 G. A. Seisenbaeva, S. Gohil, E. V. Suslova, T. V. Rogova, N. Y. Turova
and V. G. Kessler, Inorg. Chim. Acta, 2005, 358, 3506–3512.
33 D. C. Bradley, R. C. Mehrotra, I. P. Rothwell and A. Singh, Alkoxo
and Aryloxo Derivatives of Metals, Academic Press, New York, 2001,
pp. 704.
48 B. D. Murray and P. P. Power, Inorg. Chem., 1984, 23, 4584–
4588.
34 I. Krossing and A. Reisinger, Angew. Chem., Int. Ed., 2003, 42, 5725–
49 R. Becker, J. Weiss, M. Winter, K. Merz and R. A. Fischer, J. Organomet.
5728.
Chem., 2001, 630, 253–262.
35 A. Decken, J. Passmore and X. Wang, Angew. Chem., Int. Ed., 2006,
45, 2773–2777.
36 S. Welsch, B. Nohra, E. Peresypkina, V. C. Lescop, M. Scheer and R.
Re´au, Chem.–Eur. J., 2009, 15, 4685–4703.
37 I. Raabe, K. Wagner, K. Guttsche, M. Wang, M. Gratzel, G. Santiso-
Quinones and I. Krossing, Chem.–Eur. J., 2009, 15, 1966–1976.
38 D. F. Evans, J. Chem. Soc., 1959, 2003–2005.
39 S. K. Sur, J. Magn. Reson., 1989, 82, 169–173.
40 W. Seidel and K. J. Lattermann, Z. Anorg. Allg. Chem., 1982, 488,
69–74.
50 M. Hakansson, C. Lopes and S. Jagner, Inorg. Chim. Acta, 2000, 304,
178–183.
51 O. Kahn, Molecular Magnetism, VCH, 1993.
52 C. J. O’Connor, Prog. Inorg. Chem., 1982, 29, 203–283.
53 G. M. Sheldrick, Acta Crystallogr., Sect. A: Found. Crystallogr., 2008,
64, 112–122.
54 ADF2008.01, SCM, Theoretical Chemistry, Vrije Universiteit,
Amsterdam, The Netherlands, www.scm.com.
55 G. Te Velde, F. M. Bickelhaupt, E. J. Baerends, C. F. Guerra, S. J. A.
Van Gisbergen, J. G. Snijders and T. Ziegler, J. Comput. Chem., 2001,
22, 931–967.
56 C. F. Guerra, J. G. Snijders, G. Te Velde and E. J. Baerends, Theor.
Chem. Acc., 1998, 99, 391–403.
41 A. N. Kornev, T. A. Chesnokova, V. V. Semenov, E. V. Zhezlova, L. N.
Zakharov, L. G. Klapshina, G. A. Domrachev and V. S. Rusakov,
J. Organomet. Chem., 1997, 547, 113–119.
42 M. A. Putzer, B. Neumueller, K. Behnicke and J. Magull, Chem. Ber.,
1996, 129, 715–719.
57 S. H. Vosko, L. Wilk and M. Nusair, Can. J. Phys., 1980, 58, 1200–
1211.
This journal is
The Royal Society of Chemistry 2010
Dalton Trans., 2010, 39, 374–383 | 383
©