Anodic electrochemistry of Mn and Re tricarbonyl complexes of tris(oxazolinyl)phenyl borate ligands: Comparison to tris(pyrazolyl) borate complexes

Article abstract of DOI:10.1039/c1nj20222g

M(CO)₃ (M = Mn, Re) complexes of two tris(oxazolinyl) phenylborate ligands (To^M and To^P) have been prepared and characterized by spectroscopic, electrochemical and crystallographic methods. The steric bulk imparted by the sp³-hybridized carbons in the oxazoline ring gives the tris(oxazolinyl)borate ligands significant structural protection against dimerization while leaving the M(CO)₃ face open for possible reactivity. The anodic electrochemistry of these systems in relatively benign media showed that their one-electron oxidation products were stable and persistent in the case of Re, but of limited stability in the case of Mn. The anodic oxidation of Mn and Re complexes of four tris(pyrazolyl)borate (Tp and Tp*) ligands was also investigated, once again showing that the radical cations of the Re complexes were longer lived than their Mn congeners in CH₂Cl₂/[NBu₄][B(C₆F ₅)₄]. The E_1/2 data were used to classify the electron-donating properties of the different scorpionate ligands in comparison with their cyclopentadienyl counterparts as Tp* > Cp* > To^P, Tp, To^M > Cp. Comparison of the ν_CO IR frequencies of these complexes gave a slightly different sequence of donor strength, Cp*, To^M, To^P > Tp* > Tp > Cp. The electronic effects of the tris(oxazolinyl)phenyl borate ligands are that they convey "Tp*-like" character in terms of the chemical reactivity of their complexes, and "Tp-like" character in terms of the redox potentials of these complexes.

Full text of DOI:10.1039/c1nj20222g

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
NJC
Cite this: New J. Chem., 2011, 35, 2169–2178
www.rsc.org/njc
PAPER
Anodic electrochemistry of Mn and Re tricarbonyl complexes of
tris(oxazolinyl)phenyl borate ligands: comparison to tris(pyrazolyl)
borate complexeswz
Kan Wu,a Debabrata Mukherjee,b Arkady Ellern,b Aaron D. Sadow*b and
William E. Geiger*a
Received (in Victoria, Australia) 9th March 2011, Accepted 9th May 2011
DOI: 10.1039/c1nj20222g
M(CO)3 (M = Mn, Re) complexes of two tris(oxazolinyl)phenylborate ligands (ToM and ToP)
have been prepared and characterized by spectroscopic, electrochemical and crystallographic
methods. The steric bulk imparted by the sp3-hybridized carbons in the oxazoline ring gives the
tris(oxazolinyl)borate ligands significant structural protection against dimerization while leaving
the M(CO)3 face open for possible reactivity. The anodic electrochemistry of these systems in
relatively benign media showed that their one-electron oxidation products were stable and
persistent in the case of Re, but of limited stability in the case of Mn. The anodic oxidation of
Mn and Re complexes of four tris(pyrazolyl)borate (Tp and Tp*) ligands was also investigated,
once again showing that the radical cations of the Re complexes were longer lived than their Mn
congeners in CH2Cl2/[NBu4][B(C6F5)4]. The E1/2 data were used to classify the electron-donating
properties of the different scorpionate ligands in comparison with their cyclopentadienyl
counterparts as Tp* 4 Cp* 4 ToP, Tp, ToM 4 Cp. Comparison of the nCO IR frequencies of
these complexes gave a slightly different sequence of donor strength, Cp*, ToM, ToP 4 Tp* 4
Tp 4 Cp. The electronic effects of the tris(oxazolinyl)phenyl borate ligands are that they convey
‘‘Tp*-like’’ character in terms of the chemical reactivity of their complexes, and ‘‘Tp-like’’
character in terms of the redox potentials of these complexes.
three oxazoline donor groups in place of the pyrazolyl groups
of the Tp ligand class.3 The oxazoline moiety provides access
Introduction
There has been a steady interest in developing mimics of the
to chiral structures, and the substitution of B–C for B–N
cyclopentadienyl (Cp) ligand as a monoanionic formal six-
linkages limits the Lewis acid-mediated ligand epimerizations
that complicate the chemistry of metal–Tp complexes.4
electron donor. Among the most broadly investigated
analogues are the hydridotris(pyrazolyl)borate (Tp) ligands,
Owing to the fact that ligand electronic properties are
which differ from the Cp ligand in terms of steric constraints
important in determining the reactivity of metal complexes,
and metal–ligand bonding.1 A number of papers have
effort has been made to characterize the relative donor properties
compared the stoichiometric and catalytic reactivities of metal
complexes derived from these two classes of ligands.2 Quite
of Cp- vs. scorpionate-type ligands. The most comprehensive
of these studies is the compilation by Tellers et al. of the
donor-related physical parameters published prior to 2000.5
recently, one of our groups introduced a new type of scorpio-
nate ligand, tris(oxazolinyl)phenyl borate, which contains
Based largely on IR and electrochemical data, these authors
discouraged generalizations about the relative donating
a Department of Chemistry, University of Vermont, Burlington,
abilities of the two types of ligands. We took an interest in
Vermont 05405, USA. E-mail: william.geiger@uvm.edu
b Department of Chemistry and U.S. DOE Ames Laboratory,
Iowa State University, Ames IA, 50011, USA.
E-mail: sadow@iastate.edu
w Dedicated to Prof. Didier Astruc in recognition of his outstanding
contributions to organometallic electron-transfer chemistry.
z Electronic supplementary information (ESI) available: Two figures
showing cyclic voltammograms referred to in the manuscript
[(a) complex 2 before bulk electrolysis, (b) complex 7 before and after
bulk anodic electrolysis] and four sets of crystallographic data (cif) of
compounds 5–8. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/c1nj20222g
whether such a comparison could be delineated for one
particular group of complexes, namely those containing group
VII M(CO)3 moieties (M = Mn, Re). Definitive electro-
chemical E1/2 data on this set of complexes were not available
previously, owing in part to poor chemical reversibility and
sensitivity to medium effects in the oxidation of MCp(CO)3
complexes. The milder conditions possible when using weakly-
coordinating electrolyte anions6 have since allowed characteri-
zation of the simple one-electron oxidative reactions of the
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2169–2178 2169

View Article Online
ð1Þ
methylene groups of the oxazoline moiety and three downfield
multiplets corresponding to the phenyl group. The equivalence
of the oxazoline rings indicates a k3-N,N,N-coordination of
the ToM ligand, consistent with a C3v-symmetric structure.
Likewise, the 13C spectrum of 5 contained a single set of
oxazoline resonances and a single carbonyl resonance at
197.90 ppm. Crosspeaks between the oxazoline nitrogen and
the ring methyl and methylene groups in a 1H–15N HMBC
experiment (performed at natural 15N abundance) provided a
characteristic 15N NMR chemical shift of ꢀ176.8 ppm for 5.
This chemical shift is in the range of late transition-metal
complexes containing the ToM ligand and downfield from
2H-dimethyl-2-oxazoline (ꢀ127.9 ppm).
Scheme 1
MCp(CO)3 complexes, and reliable anodic E1/2 values are now
available.7 In the interest of comparing the electron-donating
abilities of cyclopentadienyl, hydridotris(pyrazolyl)borate,
and tris(oxazolinyl)phenyl borate ligands, the appropriate
M(CO)3 complexes were prepared and their anodic E1/2
potentials and IR data were measured under identical condi-
tions. Previous studies on the donor ability of oxazolines have
focused on photoelectron spectroscopy and ab initio calcula-
tion of uncoordinated groups, and few comparisons to ligands
such as cyclopentadienyl are available.8,9
One nCN band at 1582 cmꢀ1 was observed in the solid-state
infrared spectrum (KBr) of ReToM(CO)3, which further
supports the k3-coordination. The solid state IR spectrum of
ReToM(CO)3 also contains one sharp band and one broad
absorption (2012 and 1892 cmꢀ1), corresponding to the
symmetric and anti-symmetric carbonyl modes. These bands
shift to slightly higher energy in the solution-phase spectrum
acquired in THF (2013 and 1894 cmꢀ1) or dichloromethane
(2019 and 1898 cmꢀ1).
While in the process of determining the oxidation potentials
of the Tp, Tp* [Tp* = hydrotris(3,5-dimethylpyrazolyl)borate],
and oxazoline complexes, we also studied the longevity of their
one-electron oxidation products and characterized, where
possible, their IR spectra.
The structures and numerical ordering of the compounds
described in this paper are shown in Scheme 1. The acronym
ToM is used for the tris(4,4-dimethyl-2-oxazolinyl)phenyl borate
ligand and ToP is employed for the tris(4S-isopropyl-2-
oxazolinyl)phenyl borate ligand.
X-Ray quality single crystals of 5 and 7 were obtained from
a concentrated toluene solution cooled to ꢀ30 1C. The solution
to the single crystal X-ray diffraction study confirms
the identity of ReToM(CO)3 and reveals the tridentate
fac-N,N,N-ToM coordination (Fig. 1). The relatively large
rhenium(I) center and the steric and chelating properties of
the ToM ligand result in small N–Re–N angles (80.6(3)1 to
85.4(3)1). The carbonyl ligands are also slightly compressed,
with C–Re–C angles less than 901 (from 86.1(5)1 to 87.2(5)1),
away from the bulky ToM ligand. Compounds 5 and 7 form a
fine isomorphous pair.
Reaction of optically active Li[ToP] [ToP = tris(4S-isopropyl-
2-oxazolinyl)phenyl borate] and MBr(CO)5 (M = Re, Mn) in
THF at 80 1C gave MToP(CO)3 contaminated with unidenti-
fied impurities that were not removed by recrystallization from
toluene. We had previously found that the potassium salt
K[ToM] reacts under milder conditions than Li[ToM] as an
Results and discussion
Synthesis and characterization of tris(oxazolinyl)borate
compounds
Group
7
tris(oxazolinyl)borate compounds MToM(CO)3
[M = Re (5), Mn (7)] were prepared by adaptation of the salt
metathesis route that provides MTp(CO)3 (M = Mn, Re).
Thus, Li[ToM] and MBr(CO)5 (M = Mn, Re) were reacted in
THF or acetonitrile at 80 1C to afford MToM(CO)3 in good
yield (eqn (1)).
These compounds were fully characterized by spectroscopic
(Table 1), analytical, and single crystal X-ray diffraction
methods. The 1H NMR spectrum of ReToM(CO)3 (5) in
benzene-d6 contained singlet resonances for the methyl and
Table 1 Summary of the characteristic data for tris(oxazolinyl)phenylboratorhenium and manganese tricarbonyl complexes
Compound
nCO (KBr, cmꢀ1
)
nCO (solvent, cmꢀ1
)
nCN (KBr, cmꢀ1
)
15N NMR/ppm
ReToM(CO)3, 5
2012, 1892
2013, 1894 (THF)
1582
ꢀ176.8
2019, 1898 (CH2Cl2)
2012, 1896 (THF)
2014, 1897 (CH2Cl2)
2017, 1902 (THF)
2020, 1912 (CH2Cl2)
2017, 1908 (THF)
n.a.
ReToP(CO)3, 6
MnToM(CO)3, 7
2010, 1892
2018, 1899
1589
1592
ꢀ195.6
ꢀ172.1
MnToP(CO)3, 8
H[ToP]
2016, 1961
n.a.
n.a.
1601
1601 (broad)
1601
Not detected
ꢀ195.6
K[ToP]
n.a.
ꢀ149.1
c
2170 New J. Chem., 2011, 35, 2169–2178
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
effective [ToM]-transfer agent. Therefore, K[ToP] was synthe-
sized from Li[ToP] via HToP according to eqn (2). A reaction
between Li[ToP] and [HEt3N]Cl in methylene chloride, followed
by a filtration through a plug of neutral alumina, provided
anhydrous HToP in 54% yield as a light yellow gel.
K[ToP] with Re(CO)5Br and Mn(CO)5Br in THF at 80 1C
afforded ReToP(CO)3 (6) and MnToP(CO)3 (8), respectively,
in 85% and 87% isolated yields (eqn (3)). As discussed for
ReToM(CO)3, both 6 and 8 were characterized by crystal-
lography and spectroscopy (Table 1).
ð2Þ
HToP was characterized by 1H, 11B, 13C{1H}, and 15N
NMR spectroscopy, IR, and high-resolution mass spectro-
metry. In particular, the 1H NMR spectrum of HToP
in benzene-d6 shows the presence of a single diastereomer
as the S,S,S enantiomer and
a high-resolution mass
spectrum producing the parent ion peak at 426.3. HToP
reacts upon treatment with KH in THF at 25 1C to provide
K[ToP] as a light yellow solid along with hydrogen gas
evolution.
ð3Þ
X-Ray quality crystals were obtained for compounds 6 and
8, and their structures reported here are the first containing the
chiral ToP ligand. The X-ray crystal structures for these two
compounds are isomorphous. The value of the Flack para-
meter, 6 = 0.010(6) and 8 = 0.005(9), establishes the absolute
configuration of the structure as S, and this configuration is
expected from our use of L-valine to prepare 4S-isopropyl-2-
oxazoline. These structural studies provide valuable informa-
tion about the steric properties of the chiral ToP ligand as well
as the conformation favored in the solid state. The two
Although we were unable to grow X-ray quality single
crystals for K[ToP], the other solid as well as solution phase
spectroscopic data (1H, 11B, 13C, 15N NMR and IR) were
indicative of the presence of an optically active ToP pro-
ligand. In particular, the diastereotopic methyl resonances
from the isopropyl group indicate the presence of a stereo-
center, and the solution NMR data (1H, 13C, 11B, 1H–15N
HMBC) indicated a pseudo-C3-symmetric geometry of the
ToP-ligand, thus making all the three oxazoline rings equi-
valent. Thus, only one diastereomer (containing three identical
stereocenters) is present. In the solid-state IR spectrum a sharp
single band for the oxazoline CQN stretching at 1601 cmꢀ1
indicates that KToP has similar geometry as the LiToP.
As mentioned earlier, individual salt metathesis reactions of
Fig. 2 ORTEP diagram of ReToP(CO)3 (6). Hydrogen atoms on the
methine carbons are shown to illustrate the S absolute configuration
and isopropyl conformation for the oxazolines; all other hydrogen
atoms and two co-crystallized toluene molecules are not included in
the plot. A similar structure was observed for 8.
Fig. 1 ORTEP diagram of 5. Ellipsoids are drawn at 50%
probability, and hydrogen and a disordered toluene (about an
inversion center) are omitted for clarity. Compound 7 is isomorphous
with 5.
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2169–2178 2171

View Article Online
MToP(CO)3 complexes have very similar structures, and the
ORTEP diagram of 6 is shown in Fig. 2 (see ESIz for the X-ray
structure of 8). Significantly, in both 6 and 8, the isopropyl
groups are oriented so that the methine CH, rather than a
methyl group, on each oxazoline is directed toward the
Re(CO)3 moiety. This conformation appears to minimize
unfavorable inter-ligand interactions.
Table 2 Solid angles for tris(oxazolinyl)phenyl borates, tris(pyrazolyl)-
borates, and cyclopentadienyls in rhenium tricarbonyl complexes
Solid angle/
steradians
Percent of sphere
occupied
Compound
ReToM(CO)3, 5
ReToP(CO)3, 6
ReTp*(CO)3, 2
ReTp(CO)3, 1
ReCp*(CO)3, 10
ReCp(CO)3, 9
6.88
6.61
6.49
5.26
4.72
4.20
54.7
52.6
51.6
41.8
37.6
33.4
Comparison of steric properties of ToM, ToP, Tp, Tp*, Cp and
Cp* ligands in Re(CO)3 complexes
Electrochemistry of rhenium complexes
A
qualitative comparison of the steric properties of
tris(oxazolinyl)borate and tris(pyrazolyl)borate ligands was
obtained from Newman projections of the structures along
their C3 axis (Fig. 3). The steric bulk of the planar pyrazole
groups in the Tp* ligand is projected in the plane of the
aromatic ring, leaving three large wedges of open space
between the donor groups. In contrast, the non-planar oxazoline
groups in ToM and ToP give a significantly different shape.
The 4,4-dimethyl groups on ToM create a bowl-like hemi-
sphere of steric bulk around the –Re(CO)3 moiety. In ToP, the
steric bulk creates a windmill-like shape that blocks only one
side of each carbonyl ligand.
Electrochemical experiments used dried dichloromethane
containing 0.05 M [NBu4][TFAB] (TFAB = [B(C6F5)4]ꢀ) as
the supporting electrolyte, and potentials are given versus the
ferrocene/ferrocenium couple. Details are given in the Experi-
mental section. An overview of our findings is that all four
rhenium complexes showed quasi-Nernstian one-electron
oxidations giving cation radicals that were persistent, whereas
one-electron oxidation of the three manganese complexes gave
electrochemically observable, but less stable, cations.
Anodic electrochemical and spectroelectrochemical investigation
of Re complexes
Solid angles provide a quantitative assessment of steric
properties of a ligand,10 and may be calculated conveniently
using coordinates from crystal structures or molecular models
with the program Solid-G.11 A solid angle is defined as the
surface area of a projection of a ligand on a sphere surrounding
a complex, and this approach for quantifying steric properties
accounts for a ligand’s shape.
Crystal structures have been reported for ReCp(CO)3,12
ReCp*(CO)3,13 ReTp(CO)3, and ReTp*(CO)3,14 as well as
the corresponding manganese complexes. Those structures
provided the coordinates to calculate solid angles for Cp,
Cp*, Tp, and Tp* ligands (Table 2) in steradians. The crystal
structures of ToM and ToP compounds reported here also were
used to calculate solid angles. The second parameter given is
the percent of a sphere’s surface area that is taken up by the
ligand’s two dimensional projection (maximized at 100% for a
ligand that entirely surrounds a metal center).
A single feature was observed in the cyclic voltammetry (CV)
scans of the four Re complexes 1, 2, 5, and 6, of which the blue
line of Fig. 4 is representative. At room temperature, these
curves fit the standard criteria for diffusion controlled,
chemically reversible, one-electron oxidation processes. The
E1/2 values, collected in Table 3, are all negative of the
literature potential15 (1.16 V) for the oxidation of the Cp
analogue 9. It is important to note that even at relatively
high concentrations (up to 2 mM), none of the rhenium
complexes showed evidence that the radical cation reacted to
form the type of metal–metal bonded dimer dication that has
been documented in the case of radical 9+.15 The steric
hindrance of the scorpionate ligand is apparently sufficient
to mitigate against formation of the rhenium–rhenium based
dimer.
Comparison of the solid angles calculated for rhenium
compounds provides the trend in steric bulk for the ancillary
ligands: Cp o Cp* o Tp o Tp* o ToP o ToM, and the same
trend is obtained for the manganese complexes. Notably, the
tris(oxazolinyl)borate ligands occupy greater than a hemi-
sphere around the rhenium center, and thus limit possible
dimerization or ligand redistribution processes.
That the cation radicals of the rhenium complexes are long-
lived was demonstrated by bulk anodic electrolysis tracked by
IR spectroelectrochemistry. Bulk electrolyses of 1, 2, 5 and 6
were performed at room temperature employing a platinum
gauze electrode at an applied potential which was at least
200 mV positive of the compound’s E1/2 value. Both 1 and 2
released ca. 1.0 F after complete oxidation as the solutions
Fig. 3 ORTEP diagrams of ReTp*(CO)3,14 ReToM(CO)3, and ReToP(CO)3 molecules viewed along the C3-axis.
c
2172 New J. Chem., 2011, 35, 2169–2178
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Fig. 4 Blue line is CV scan of 0.95 mM 2 in CH2Cl2/0.05 M [NBu4][TFAB] prior to bulk electrolysis. After bulk anodic electrolysis at Eappl
=
1.25 V, an identical scan initiated at 1.25 V produced the CV having the red line owing to formation of 2+. Conditions: 298 K, 1 mm GCE,
200 mV sꢀ1. A CV of the neutral complex alone is available in ESIz as Fig. S1.
Table 3 Summary of E1/2 and IR data for compounds in CH2Cl2 at room temperature, except for 5+ and 6+, which were obtained at 253 K.
0.05 M [NBu4][B(C6F5)4] was present in measurements of cationic species
Avg nco (cmꢀ1) =
(nsym + 2nasym)/3
sym/cmꢀ1
n
asym/cmꢀ1
0/+
Cmpd
E1/2 (V vs. FeCp2
)
n
ReTp(CO)3, 1
1+
ReTp*(CO)3, 2
2+
0.98
2026
2097
2018
2079
2035
2027
2019
2086
2014
2085
2020
2024
2105
2005
2062
2022
2116
2004
2093
1912
2021, 1986
1903
2002, 1969
1932
1922
1898
2005, 1962
1897
2004, 1964
1912
1926
2031
1906
2015
1934
2061, 2050
1917
2033
1950
2035
1941
2017
1966
1957
1938
2018
1936
2018
1948
1959
2056
1939
2031
1963
2078
1946
2053
0.82
MnTp(CO)3, 3
MnTp*(CO)3, 4
ReToM(CO)3, 5
5+
0.74
0.59
1.00
ReToP(CO)3, 6
6+
0.97
MnToM(CO)3, 7
ReCp(CO)3, 9
0.72
1.16
2+
92
ReCp*(CO)3, 10
0.91
0.92
0.64
2+
102
MnCp(CO)3, 11
11+
MnCp*(CO)3, 12
12+
turned deep purple. CV scans on the electrolyzed solutions
showed only one product wave (see Fig. 4), a reversible
reduction attributed to 2+ (or 1+). Electrolyses of 5 and 6
at 298 K (1.2 F passed) produced deep green solutions
of the monocations 5+ and 6+. Subsequent voltammetry
and back-electrolysis showed that some decomposition of
the cations had occurred. However, when the electrolyses were
carried out at 253 K, 5+ and 6+ were more stable and
back-electrolysis at Eappl = 0.40 V regenerated about 90%
of the starting material (based on CV and linear scan voltam-
metry data). Thus, the oxidation of all four rhenium scorpio-
nates follows the simple one-electron reaction of eqn (4). In
none of
these cases were side-product waves observed giving evidence
of the formation of hydrogen or chlorine atom-transfer
products, [Re(CO)3(scorpionate)X]+ (X = H or Cl), such as
were found for the Cp analogue.15 This may be due to the
milder oxidizing strength of the metal scorpionate cations.
The Re complexes were also investigated by fiber-optic
spectroelectrochemical experiments recorded in the metal
carbonyl IR region. The example of compound 2 at room
temperature is given in Fig. 5. As the electrolysis proceeded,
the bands at 2018 and about 1903 cmꢀ1 for neutral 2 were
replaced by the three new bands of 2+ at 2079, 2002, and
1969 cmꢀ1. Similar results were obtained for the other rhenium
complexes, with the electrolyses being carried out at 253 K for
compounds 5 and 6 to accommodate the shorter lifetimes of
the cation radicals derived from the ToM and ToP complexes.
Re(CO)3(scorpionate) " [Re(CO)3(scorpionate)]+ + eꢀ (4)
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2169–2178 2173

View Article Online
Fig. 5 Representative fiber-optic IR spectra of 1.5 mM 2 in CH2Cl2/0.05 M [NBu4][TFAB] as the bulk oxidation proceeded at 298 K with Eappl
1.0 V (vs. FeCp20/+). Spectra of 5 and 6 (neither shown here) were obtained at 253 K.
=
The IR data are collected in Table 3. It is worth noting that all
three carbonyl bands are resolved in the spectra of the radical
cations. Whether the implied reduction in C3v symmetry arises
from structural changes or from medium effects is not known.
Shifts in the carbonyl IR frequencies were evaluated as
a weighted average of the three nCO absorptions expressed
in cmꢀ1. This takes into account that two or three CO bands
may be observed, depending on whether or not the formally
degenerate pair of asymmetric bands is resolved. Thus, the
average value of nCO is the simple average if all three bands are
resolved, but takes on the value [nCO(sym) + 2nCO(asym)]/3 if the
asymmetric bands are not resolved. The average increase in
CO frequencies is between 76 cmꢀ1 and 85 cmꢀ1 in going from
the neutral complexes to the radical cations. These shifts are a
bit small compared to those reported for a number of other
piano-stool metal carbonyl oxidation processes, which tend to
temperature. Our results suggest that the radical cations would
be NMR-active in room temperature solutions, but experi-
ments of this type were not attempted.
Electrochemistry of manganese complexes
The manganese complexes 3, 4 and 7 all gave one-electron
oxidations at somewhat milder potentials than the corres-
ponding Re complexes, consistent with the behavior of their
Cp analogues.7,15 The corresponding radical cations were less
stable, however, than their Re counterparts. Thus, a CV scan
rate of 0.2 V sꢀ1 showed some degree of anodic chemical
irreversibility for each of the complexes, and the oxidation of
complex 7 was fully irreversible at room temperature. At lower
temperatures, greater chemical reversibility was observed in
each case (see Fig. 6). From the reverse-to-forward current
ratios at different CV scan rates, we estimate the half-lives of
the Mn-based radicals at room temperature to decrease in the
order 4+ (2.2 s) 4 3+ (0.8 s) 4 7+ (o0.4 s).19 Similarly, the
half-life of 7+ was determined to be 8 s at 253 K. Bulk anodic
electrolysis of each of these complexes at 253 K confirmed the
one-electron process, as 1.0 F was passed in each case, with
only 10% to 30% yields of the radical cations. No other
electroactive products were detected. Attempts at determining
IR spectra of the radical cations gave uncertain results.
Rationalization of the shorter lifetimes of the Mn cations
compared to their Re counterparts must be somewhat
speculative, owing to the fact that the follow-up reaction
products of the former have not been determined. Since bulk
electrolyses of these complexes did not yield electroactive
follow-up products, gross decomposition may be occurring.
In that case, weaker metal-to-scorpionate bonding for Mn
compared to Re may be responsible for the difference in
radical longevity. It is important to emphasize that the E1/2
potentials measured for both the Re and Mn complexes are
thermodynamically significant.
be greater than 100 cmꢀ1 7,16
.
However, a direct comparison with the ReCp complex 9 is not
available owing to the dominant dimerization reaction of its
radical cation. The CO weighted-average for the metal–metal
bonded dimer dication [Re2Cp2(CO)6]2+ is 97 cmꢀ1 and that of
its Cp* analogue is 92 cmꢀ1 15
It is safe to conclude that
.
oxidation of the Re(CO)3(scorpionate) complexes involves a
predominantly metal-based orbital, but it would not be surpris-
ing if calculations were to show that the degree of metal character
in the SOMO of the radical cation was somewhat less than that
seen for many other 17-electron half-sandwich complexes.17
Attempts were made to record ESR spectra of the cation
radicals. Solutions containing them were taken from electro-
lysis solutions, transferred to an ESR sample tube, and frozen
to 77 K before removing the samples from the drybox. Each of
the radical cations was ESR-silent at 77 K, owing undoubtedly
to fast electronic relaxation effects, in concert with the
behavior of a large number of d5 sandwich or half-sandwich
radicals possessing pseudo-C3v or higher symmetry.7,18 These
solutions were also, as expected, ESR-silent at room
c
2174 New J. Chem., 2011, 35, 2169–2178
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
Fig. 6 Representative CVs of 1.3 mM 4 in CH2Cl2/0.05 M [NBu4][TFAB] at 298 K (red) and 253 K (blue), 200 mV sꢀ1, 1.5 mm GCE. The anodic
peak current becomes smaller at lower temperature due to the smaller diffusion coefficient.
Although the E1/2 values were measured with an accuracy of
Analysis and conclusions
better than 10 mV, differences of more than 30 mV were
Combining the present results with literature values allows
judged to be necessary to place a ligand in a different rank.
comparison of the spectroscopic IR and electrochemical E1/2
The most commonly employed quantitative measure of
data for a dozen scorpionate and cyclopentadienyl complexes
ligand E1/2 shifts is the ‘‘ligand electronic factor’’ EL popular-
ized by Lever.24 The EL values of the scorpionate ligands can
of M(CO)3, M = Mn or Re. In terms of the IR data, ligand
(L) donor properties have traditionally been ranked by the nCO
be easily obtained through the E1/2 shifts (in volt) of their
frequencies of ML(CO) complexes, with decreasing stretching
complexes from those of the corresponding MCp(CO)3
frequencies being interpreted as increasing
L
donor
complexes, since the EL value of the Cp ligand (0.33) is
known.24 The EL values for the six relevant ligands are
collected in Table 5. By this measure, the donor strengths of
the tris(oxazolinyl)phenylborate ligands are essentially the
same as the Tp ligand but, in contrast to the conclusions
based on IR data, less than the Tp* ligand (see Table 4).
The apparent dichotomy between the IR and electrochemical-
based conclusions may arise from an inherent difference
between the two analytical approaches: the IR absorption
energies are determined by the atomic motions of the molecule
in one redox state, whereas the E1/2 value is determined by the
strength.4,20 Based on the weighted average of the three nCO
frequencies listed in Table 3, the scorpionate and cyclopenta-
dienyl donor properties fall in the sequence Cp*, ToM, ToP 4
Tp* 4 Tp, Cp for the Mn complexes (Table 4). In this, we
make no distinction in rank unless the difference in averaged
frequencies is greater than the resolution of the IR measure-
ment (4 cmꢀ1). For the Re complexes, the sequence is little
changed: Cp*, ToM, ToP 4 Tp* 4 Tp 4 Cp. This analysis
follows almost exactly the earlier trends summarized by Tellers
5
et al. for Tp- and Cp-type complexes and suggests that the
tris(oxazolinyl)phenylborate ligands have an increased donor
strength compared to Tp*.
Table 5 Literature EL values in this paper are taken from: aref. 24,
c
d
bref. 26, ref. 7 or ref. 15(b)
A second commonly-used measure of ligand donor strength
is the shift in E1/2 for the one-electron oxidation of metal
complexes (ML0L) in which only L is changed. This approach
has been employed and discussed in a number of different
contexts,21–24 the underlying assumption being that increasing
ligand donor strength is manifested in decreasing E1/2 values.
In this analysis, the donor strength for the six ligands under
examination decreases in the sequence Tp* 4 Cp* 4 ToP, Tp,
ToM 4 Cp (Table 4) for both Mn and Re complexes.25
EL (this work)
of Mn, Re cmpds
EL (avg,
this work)
Ligand
EL (literature)
Cp
0.33a
0.33a
0.33a
0.15
0.15
0.14
0.065
ꢀ0.005
ToM
Tp
Not reported
0.42b (see the text)
Not reported
0.06a
0.17, 0.13
0.15, 0.15
0.14 (Re)
0.08,c 0.05d
0.00, ꢀ0.01
ToP
Cp*
Tp*
c,d
Not reported
Table 4 Rankings of ligand donor properties based on data reported in this paper. The average IR frequency is the weighted average of the three
CO stretching frequencies in dichloromethane, taken from data in Table 3
Experimental basis
Rankings of ligand donor property
Avg nCO of Mn complexes 3, 4, 7, 11, 12
Avg nCO of Re complexes 1, 2, 5, 6, 9, 10
Anodic E1/2 of Mn complexes 3, 4, 7, 11, 12
Anodic E1/2 of Re complexes 1, 2, 5, 6, 9, 10
Cp*, ToM, ToP 4 Tp* 4 Tp, Cp
Cp*, ToM, ToP 4 Tp* 4 Tp 4 Cp
Tp* 4 Cp* 4 ToP, Tp, ToM 4 Cp
Tp* 4 Cp* 4 ToP, Tp, ToM 4 Cp
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2169–2178 2175

View Article Online
difference in free energy for the molecule in two redox states.
The E1/2-based analysis assumes that the energetics of
any changes in medium effects and molecular structure are
constant within the compounds being compared. The rigidity
of the pseudo-octahedral complexes under investigation would
seem to eliminate concerns about significant structural
rearrangements. In terms of the possibility of significant
differences in medium effects for redox reactions involving
the six types of ligands, the E1/2 values were measured under
relatively benign conditions in which changes in both solvation
and ion-pairing effects would be minimized.27 However, we
cannot rule out differences in medium effects that could
influence the E1/2 values. A reasonable working model for
the electronic effects of the tris(oxazolinyl)phenyl borate
ligands is that they convey ‘‘Tp*-like’’ character in terms of
the chemical reactivity of their complexes, and ‘‘Tp-like’’
character in terms of the redox potentials of their complexes.
We also point out that the EL value determined hereby for
the Tp ligand (0.14) is considerably less than the value of
0.42 previously reported based on electrochemical measure-
ments on RuTpClL2 complexes, where L is a strongly
donating phosphine.26 The source of this discrepancy may
arise from the dramatic difference in the properties of the
ancillary ligands in the M(CO)3 and RuClL2 systems.
prepared by metathesis of [NBu4]Br with K[B(C6F5)4] in
MeOH,33 recrystallized three times from CH2Cl2/OEt2, and
dried at 90 1C overnight.
ReToM(CO)3 (5)
A solution of LiToM (0.419 g, 1.08 mmol) in THF (20 mL) was
added to Re(CO)5Br (0.439 g, 1.08 mmol) suspended in THF
(5 mL). The solution mixture was degassed with freeze–
pump–thaw cycles (3ꢂ) and then heated to 80 1C for 8 h in
a sealed flask. The volatile materials were removed under
reduced pressure giving a yellowish solid, which was then
extracted with benzene (12 mL). The benzene was evaporated
to dryness providing an off-white solid that was washed with
pentane (3 ꢂ 5 mL) and further dried under vacuum yielding
0.635 g of ReToM(CO)3 (0.973 mmol, 90.0%) as a white
powder. X-Ray quality single crystals were grown from a
1
concentrated toluene solution of ReToM(CO)3 at ꢀ30 1C. H
NMR (400 MHz, methylene chloride-d2): d 1.08 (s, 18H,
CNCMe2CH2O), 3.36 (s, 6H, CNCMe2CH2O), 7.36 (t, JHH =
3
3
6.6 Hz, 1H, para-C6H5), 7.54 (t, JHH = 7.6 Hz, 2H,
3
meta-C6H5), 8.28 (d, JHH = 7.2 Hz, 2H, ortho-C6H5).
13C{1H} NMR (175 HMz, methylene chloride-d2): d 28.18
(CNCMe2CH2O), 70.47 (CNCMe2CH2O), 81.15 (CNCMe2-
CH2O), 126.16 (para-C6H5), 126.95 (meta-C6H5), 135.77 (ortho-
C6H5), 145.71 (br, ipso-C6H5), 188.43 (br, CNCMe2CH2O),
197.90 (Re(CO)3). 11B NMR (128 MHz, methylene chloride-d2):
d ꢀ17.4. 15N{1H} NMR (71 MHz, methylene chloride-d2):
d ꢀ176.8. IR (KBr, cmꢀ1): 2973 (w), 2934 (w), 2012 (s, CO),
1892 (s, CO), 1582 (s, CN), 1462 (m), 1390 (w), 1373 (w), 1354
(w), 1289 (m), 1254 (w), 1203 (m), 1179 (w), 1161 (w), 963 (w).
Anal. calcd. for C24H29BN3O6Re: C, 44.18; H, 4.48; N, 6.44.
Found: C, 44.13; H, 4.61; N, 6.07. Mp: 236 1C (decomp.).
Finally, the persistence (i.e., thermal stability) of the Re and
Mn-based cation radicals [ML(CO)3]+ under the present
electrolyte conditions is ranked in the sequences (for ReL):
ReTp, ReTp* 4 ReToM, ReToP 4 ReCp, ReCp*28 and
(for MnL): MnCp, MnCp* c MnTp* 4 MnTp 4 MnToM.
Experimental
All reactions were performed under a dry argon atmosphere
using standard Schlenk techniques or under a nitrogen atmo-
sphere in a glovebox, unless otherwise indicated. Water and
oxygen were removed from benzene, toluene, pentane, aceto-
nitrile, diethyl ether, and tetrahydrofuran solvents using an IT
PureSolv system at Iowa State or using previously described
methods29 at Vermont. Benzene-d6 and tetrahydrofuran-d8
were heated to reflux over Na/K alloy and vacuum-transferred.
Re(CO)5Br and Mn(CO)5Br were prepared according to the
literature procedure,30 starting from Re2(CO)10 and Mn2(CO)10
respectively. Re(CO)5Br was further purified by sublimation.
Mn(CO)5Br was used without further sublimation. Li[ToM]3(a)
and Li[ToP]3(c) were prepared by literature procedures. 15N
chemical shifts were determined by 1H–15N HMBC experiments
on a Bruker Avance II 700 spectrometer with a Bruker
Z-gradient inverse TXI 1H/13C/15N 5 mm cryoprobe; 15N
chemical shifts were originally referenced to liquid NH3 and
recalculated to the CH3NO2 chemical shift scale by adding
ꢀ381.9 ppm. 11B NMR spectra were referenced to an external
sample of BF3ꢁEt2O. Elemental analyses were performed using
a Perkin-Elmer 2400 Series II CHN/S by the Iowa State
Chemical Instrumentation Facility or at the Robertson laboratory
for samples prepared in Vermont. X-Ray diffraction data were
collected on a Bruker-AXS SMART or APEX II as described
in crystallographic files (cif) as part of the ESI.z The Tp and
Tp* complexes of Mn and Re were prepared by the literature
methods.31,32 The supporting electrolyte, [NBu4][B(C6F5)4], was
MnToM(CO)3 (7)
The preparation of MnToM(CO)3 follows the method used for
ReToM(CO)3 given above. LiToM (0.380 g, 0.976 mmol) and
Mn(CO)5Br (0.269 g, 0.979 mmol) afforded 0.454 g of
MnToM(CO)3 (0.871 mmol, 89.0%) as a crystalline yellow
solid following that procedure. X-Ray quality single crystals
were grown from a concentrated toluene solution of MnToM(CO)3
1
at ꢀ30 1C. H NMR (400 MHz, benzene-d6): d 1.16 (s, 18H,
CNCMe2CH2O), 3.40 (s, 6H, CNCMe2CH2O), 7.35 (br m,
1H, para-C6H5), 7.53 (br m, 2H, meta-C6H5), 8.30 (br m, 2H,
ortho-C6H5). 13C{1H} NMR (175 HMz, benzene-d6): d 27.53
(CNCMe2CH2O), 69.54 (CNCMe2CH2O), 80.96 (CNCMe2-
CH2O), 126.30 (para-C6H5), 127.29 (meta-C6H5), 136.32
(ortho-C6H5), 142.80 (br, ipso-C6H5), 189.59 (br, CNCMe2-
CH2O), 222.74 (Mn(CO)3). 11B NMR (128 MHz, benzene-d6):
d ꢀ17.1. 15N{1H} NMR (71 MHz, benzene-d6): d ꢀ172.1. IR
(KBr, cmꢀ1): 2969 (w), 2930 (w), 2879 (w), 2018 (s, CO), 1899
(s, CO), 1592 (s, CN), 1463 (m), 1390 (w), 1372 (w), 1352 (w),
1285 (m), 1200 (m), 1181 (w), 1158 (w), 993 (w), 969 (w). Anal.
calcd. for C24H29BN3O6Mn: C, 55.30; H, 5.61; N, 8.06.
Found: C, 55.10; H, 5.75; N, 8.05. Mp: 210–212 1C.
HToP
LiToP (6.150 g, 14.26 mmol) and [HNEt3]Cl (2.159 g, 15.68 mmol)
were dissolved in methylene chloride and were stirred for 8 h at
c
2176 New J. Chem., 2011, 35, 2169–2178
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011

View Article Online
3
room temperature. The solution was filtered through a plug
of neutral alumina. The methylene chloride was evaporated
(acetonitrile-d3, 400 MHz): d 0.64 (d, 9H, JHH = 4.0 Hz,
3
CNC(CHMe2)HCH2O), 0.94 (d, 9H, JHH
= 4.0 Hz,
under reduced pressure giving
a
yellow solid, which
CNC(CHMe2)HCH2O), 2.46 (m, 3H, CNC(CHMe2)HCH2O),
4.1 (m, 3H, CNC(CHMe2)HCH2O), 4.26 (m, 6H,
was extracted with benzene. The filtrate was evaporated to
dryness, providing 3.245 g of HToP (7.63 mmol, 53.5%)
3
CNC(CHMe2)HCH2O), 7.13 (t, 1H, JHH = 4.2 Hz, para-
as a yellow gel. 1H NMR (benzene-d6, 400 MHz): d 0.68
C6H5), 7.19 (t, 2H, JHH = 4.4 Hz, meta-C6H5), 7.63 (d, 2H,
3JHH = 4.0 Hz, ortho-C6H5). 13C{1H} NMR (acetonitrile-d3,
3
3
(d, 9H, JHH
=
6.8 Hz, CNC(CHMe2)HCH2O), 0.87
3
(d, 9H, JHH = 6.4 Hz, CNC(CHMe2)HCH2O), 1.43 (m, 3H,
100 MHz):
d
13.89 (CNC(CHMe2)HCH2O), 18.90
CNC(CHMe2)HCH2O), 3.44 (m, 3H, CNC(CHMe2)HCH2O),
3
3.61 (t, 3H, JHH = 8.4 Hz, CNC(CHMe2)HCH2O), 3.85
(CNC(CHMe2)HCH2O), 30.19 (CNC(CHMe2)HCH2O),
70.63 (CNC(CHMe2)HCH2O), 73.83 (CNC(CHMe2)HCH2O),
126.54 (para-C6H5), 127.43 (meta-C6H5), 136.16 (ortho-C6H5),
142.62 (br, ipso-C6H5), 189.35 (br, CNCMe2CH2O), 198.31
(Re(CO)3). 11B NMR (acetonitrile-d3, 128.4 MHz): d ꢀ18.1.
15N NMR (acetonitrile-d3, 71 MHz): d ꢀ195.6. IR (KBr, cmꢀ1):
3045 (w), 2964 (m), 2873 (w), 2010 (s, CO), 1892 (s, br, CO),
1589 (s, CN), 1480 (m), 1463 (m), 1391 (m), 1373 (m), 1363
(m), 1352 (m), 1324 (w), 1277 (w), 1219 (s), 1143 (w), 1115 (m),
1046 (w), 1016 (w), 971 (m), 956 (m), 874 (m), 795 (w), 739 (w),
702 (m). Anal. calcd. for C247H35BN3O6Re: C, 46.69; H, 5.08;
N, 6.05. Found: C, 45.94; H, 4.93; N, 5.82. Mp: 205–207 1C
(decomp.).
3
(m, 3H, CNC(CHMe2)HCH2O), 7.26 (t, 1H, JHH = 7.2 Hz,
3
para-C6H5), 7.46 (t, 2H, JHH = 7.2 Hz, meta-C6H5), 8.11
(br, 2H, ortho-C6H5). 13C{1H} NMR (benzene-d6, 100 MHz):
d 19.00 (CNC(CHMe2)HCH2O), 19.01 (CNC(CHMe2)HCH2O),
33.20 (CNC(CHMe2)HCH2O), 69.56 (CNC(CHMe2)HCH2O),
71.06 (CNC(CHMe2)HCH2O), 126.40 (para-C6H5), 127.87
(meta-C6H5), 134.90 (ortho-C6H5), 146.20 (br, ipso-C6H5),
188.10 (br, CNCMe2CH2O). 11B NMR (benzene-d6,
128.4 MHz): d ꢀ16.6. 15N NMR (benzene-d6, 71 MHz):
d ꢀ195.6. IR (KBr, cmꢀ1): 2961 (w, br), 1601 (s, CN), 1468
(m), 1424 (w), 1385 (w), 1313 (w), 1262 (m), 1169 (w), 1105 (w),
1026 (w), 968 (m). MS (TOF EI) Exact mass calc. for
C24H36BN3O3: m/e 425.2850 ([M]+), found 425.2866.
MnToP(CO)3 (8)
The preparation of MnToP(CO)3 follows the method used for
ToMRe(CO)3 given above. KToP (0.213 g, 0.460 mmol)
and Mn(CO)5Br (0.127 g, 0.462 mmol) yielded 0.227 g of
MnToP(CO)3 (0.402 mmol, 87%) as a yellow solid. X-Ray
quality single crystals were grown from a concentrated
solution of MnToP(CO)3 in a benzene–pentane mixture at
KToP
Solid KH was slowly added in small portions to a solution of
HToP (0.602 g, 1.415 mmol) in THF (15 mL) until gas
evolution was no longer observed. The suspension was stirred
at room temperature for 1 h and filtered. THF was evaporated
affording a pale yellow gel, which was then triturated with
pentane providing 0.616 g of KToP (1.33 mmol, 93.9%) as an
off-white solid. 1H NMR (acetonitrile-d3, 400 MHz): d 0.79
1
ambient temperature. H NMR (acetonitrile-d3, 400 MHz): d
0.64 (d, 9H, JHH = 4.0 Hz, CNC(CHMe2)HCH2O), 0.94
3
(d, 9H, 3JHH = 4.0 Hz, CNC(CHMe2)HCH2O), 2.46 (m, 3H,
3
(d, 9H, JHH = 6.8 Hz, CNC(CHMe2)HCH2O), 0.90 (d, 9H,
3JHH = 6.8 Hz, CNC(CHMe2)HCH2O), 1.58 (m, 3H,
3
CNC(CHMe2)HCH2O), 3.54 (t, 3H, JHH
CNC(CHMe2)HCH2O), 4.1 (m, 3H, CNC(CHMe2)HCH2O),
3
4.26 (m, 6H, CNC(CHMe2)HCH2O), 7.13 (t, 1H, JHH
=
3
4.2 Hz, para-C6H5), 7.19 (t, 2H, JHH = 4.4 Hz, meta-C6H5),
=
7.6 Hz,
3
CNC(CHMe2)HCH2O), 3.69 (m, 3H, CNC(CHMe2)HCH2O),
3
3.79 (m, 3H, CNC(CHMe2)HCH2O), 6.94 (t, 1H, JHH
7.63 (d, 2H, JHH = 4.0 Hz, ortho-C6H5). 13C{1H} NMR
=
3
6.8 Hz, para-C6H5), 7.02 (t, 2H, JHH = 7.2 Hz, meta-C6H5),
(benzene-d6, 100 MHz): d 13.96 (CNC(CHMe2)HCH2O),
19.17 (CNC(CHMe2)HCH2O), 29.90 (CNC(CHMe2)HCH2O),
69.90 (CNC(CHMe2)HCH2O), 72.95 (CNC(CHMe2)HCH2O),
126.17 (para-C6H5), 127.22 (meta-C6H5), 136.03 (ortho-C6H5),
142.80 (br, ipso-C6H5), 189.43 (br, CNCMe2CH2O), 222.85
(Mn(CO)3). 11B NMR (benzene-d6, 128.4 MHz): d ꢀ18.1. IR
(KBr, cmꢀ1): 3045 (w), 2963 (m), 2873 (w), 2016 (s, CO), 1961
(s, br, CO), 1601 (s, CN), 1496 (w), 1481 (m), 1464 (m), 1432
(w), 1391 (m), 1372 (m), 1364 (m), 1350 (m), 1325 (w), 1278
(w), 1226 (s), 1116 (m), 1044 (m), 972 (m), 873 (m), 798 (m),
738 (w), 702 (m). Anal. calcd. for C27H35BN3O6Mn: C, 57.57;
H, 6.26; N, 7.46. Found: C, 57.20; H, 6.53; N, 7.36. Mp:
194–196 1C.
7.56 (d, 2H, JHH = 6.4 Hz, ortho-C6H5). 13C{1H} NMR
3
(acetonitrile-d3, 100 MHz): d 18.35 (CNC(CHMe2)HCH2O),
19.88 (CNC(CHMe2)HCH2O), 33.67 (CNC(CHMe2)HCH2O),
67.29 (CNC(CHMe2)HCH2O), 73.72 (CNC(CHMe2)HCH2O),
124.80 (para-C6H5), 126.90 (meta-C6H5), 135.90 (ortho-C6H5),
152.74 (br, ipso-C6H5), 183.65 (br, CNCMe2CH2O). 11B NMR
(benzene-d6, 128.4 MHz): d ꢀ16.7. 15N NMR (benzene-d6,
71 MHz): d ꢀ149.1. IR (KBr, cmꢀ1): 2956 (w, br), 2882 (w),
1601 (s, CN), 1480 (m), 1467 (m), 1430 (m), 1385 (m), 1367
(m), 1263 (m), 1105 (w, br), 967 (m, br), 899 (m), 855 (m), 739
(w), 705 (m). Exact mass calc. for C48H70B2KN6O6: m/e
887.5180 ([2M-K]+), found 887.5279. Mp: 95–97 1C.
Electrochemistry
ReToP(CO)3 (6)
Except for IR spectroelectrochemistry which was conducted
under Schlenk-type conditions, all electrochemistry was
carried out in a Vacuum Atmospheres drybox under nitrogen,
using a Princeton Applied Research (PAR) Model 273A
potentiostat interfaced to a personal computer. Voltammetry
scans were recorded using a glassy carbon working electrode
disk of either 1 or 3 mm diameter (Bioanalytical Systems).
The preparation of ReToP(CO)3 follows the method used for
ReToM(CO)3 given above. KToP (0.314 g, 0.677 mmol) and
Re(CO)5Br (0.276 g, 0.677 mmol) afforded 0.400 g of ReToP(CO)3
(0.576 mmol, 85.1%) as a pale yellow solid. X-Ray quality
colorless single crystals were grown from
trated toluene solution of ReToP(CO)3 at ꢀ30 1C. H NMR
a concen-
1
c
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011
New J. Chem., 2011, 35, 2169–2178 2177

View Article Online
The disks were pretreated using a standard sequence of
polishing with diamond paste (Buehler) of decreasing sizes
(3 to 0.25 mm) interspersed by washings with nanopure water
and finally vacuum drying. Bulk electrolyses were carried out
in a three-compartment ‘‘H-type’’ cell having reference, counter
and working compartments separated by two fine glass frits
and the working electrode was a basket shaped Pt gauze. The
10 D. White and N. J. Coville, Adv. Organomet. Chem., Academic
Press, New York, 1994, p. 95.
11 (a) I. A. Guzei and M. Wendt, Dalton Trans., 2006, 3991;
(b) I. A. Guzei and M. Wendt, Program Solid-G, UW-Madison,
WI, USA, 2004.
12 P. J. Fitzpatrick, Y. Le Page and I. S. Butler, Acta Crystallogr.,
Sect. B: Struct. Crystallogr. Cryst. Chem., 1981, 37, 1052.
13 K. Raptis, E. Dornberger, B. Kanellakopulos, B. Nuber and
M. L. Ziegler, J. Organomet. Chem., 1991, 408, 61.
reference electrode in all cases was a Ag/AgCl wire, but all
0/+
14 J. E. Joachim, C. Apostolidis, B. Kanellakopulos, R. Maier,
N. Marques, D. Meyer, J. Muller, A. Pires de Matos, B. Nuber,
¨
potentials have been recorded versus the FeCp2
redox
J. Rebizant and M. L. Ziegler, J. Organomet. Chem., 1993,
448, 119.
couple obtained by using FeCp2 as an internal standard.34
IR spectra at Vermont were obtained with an ATI-Mattson
Infinity Series FTIR interfaced to a personal computer
employing Winfirst software at a resolution of 4 cmꢀ1. In situ
spectroelectrochemistry was performed using a mid-IR fiber-
optic ‘‘dip’’ probe (Remspec, Inc) in conjunction with the
Mattson FTIR,35 and a standard H-type electrolysis cell under
15 Re complex: (a) D. Chong, A. Nafady, P. J. Costa, M. J. Calhorda
and W. E. Geiger, J. Am. Chem. Soc., 2005, 127, 15676;
(b) D. Chong, D. R. Laws, A. Nafady, P. J. Costa, A. L.
Rheingold, M. J. Calhorda and W. E. Geiger, J. Am. Chem.
Soc., 2008, 130, 2692.
16 A weighted average shift of +115 cmꢀ1 was observed for
[Cr(CO)3(Z6-C6H6)]0/+: (a) N. C. Ohrenberg, L. M. Paradee,
R. J. DeWitte III, D. Chong and W. E. Geiger, Organometallics,
2010, 29, 3179; (b) N. Camire, A. Nafady and W. E. Geiger, J. Am.
Chem. Soc., 2002, 124, 7260. The same shift was observed for
[Mn(CO)3(Z5-C5H5)]0/+ (see ref. 7).
1
argon. H NMR spectra were obtained with a Bruker ARX
500 MHz spectrometer, and ESR spectra were recorded using
a Bruker ESP 300E spectrometer.
17 It is tempting to interpret smaller carbonyl IR shifts in the
oxidations of half-sandwich carbonyl complexes with decreased
metal character of the HOMO. However, the breadth of this
correlation has not been established, and more systematic work
in this area would be welcome. Average nCO shifts of 107–115 cmꢀ1
have been reported for metal tricarbonyl and metal dicarbonyl
phosphine derivatives in which the apparent % metal in the
HOMO varies between 50% and 80–90%. In addition to ref. 7
and 16, see (a) M. P. Castellani, N. G. Connelly, R. D. Pike,
A. L. Rieger and P. H. Rieger, Organometallics, 1997, 16, 4369;
(b) R. D. Pike, A. L. Rieger and P. H. Rieger, J. Chem. Soc.,
Faraday Trans. 1, 1989, 85, 3913.
18 P. H. Rieger, Coord. Chem. Rev., 1994, 135–136, 203.
19 A first-order decomposition was assumed, with the follow-up rate
constantly being determined through the use of the working curve
found in R. S. Nicholson and I. Shain, Anal. Chem., 1965, 37, 179.
20 (a) D. R. Anton and R. H. Crabtree, Organometallics, 1983, 2, 621;
(b) C. A. Tolman, J. Am. Chem. Soc., 1970, 92, 2953.
Acknowledgements
We thank Dr D. Bruce Fulton for assistance with 15N NMR
measurements. This work was supported at Vermont by the
National Science Foundation (CHE-0808909) and at Iowa
State University by the U.S. Department of Energy, Office
of Basic Energy Sciences, Division of Chemical Sciences,
Geosciences, and Biosciences through the Ames Laboratory
(Contract No. DE-AC02-07CH11358). Aaron D. Sadow is an
Alfred P. Sloan Fellow.
References
1 (a) S. Trofimenko, Scorpionates—The Coordination Chemistry of
Polypyrazolylborate Ligands, Imperial College Press, London,
1999; (b) S. Trofimenko, Chem. Rev., 1993, 93, 943; (c) J. Smith,
Comments Inorg. Chem., 2008, 29, 189; (d) N. J. Beach,
A. E. Williamson and G. J. Spivak, J. Organomet. Chem., 2005,
690, 4640.
2 (a) W. D. Jones and F. J. Feher, J. Am. Chem. Soc., 1982,
104, 4240; (b) D. M. Tellers and R. D. Bergman, J. Am. Chem.
Soc., 2001, 123, 11508; (c) R. D. Bergman, T. R. Cundari,
T. B. Gunnoe, W. D. Harman, T. R. Klinckman, M. D. Temple
and M. D. White, Organometallics, 2003, 22, 2331; (d) S. Milione,
C. Cuomo and A. Grassi, Top. Catal., 2006, 40, 163.
3 (a) J. F. Dunne, J. Su, A. Ellern and A. D. Sadow, Organometallics,
2008, 27, 2399; (b) H.-A. Ho, J. F. Dunne, A. Ellern and A. D. Sadow,
Organometallics, 2010, 29, 4105; (c) B. Baird, A. V. Pawlikowski, J. Su,
J. W. Wiench, M. Pruski and A. D. Sadow, Inorg. Chem., 2008,
47, 10208.
21 (a) C. J. Pickett and D. Pletcher, J. Organomet. Chem., 1975,
102, 327; (b) J. Chatt, C. T. Kan, C. J. Leigh, C. J. Pickett and
D. R. Stanley, J. Chem. Soc., Dalton Trans., 1980, 2032.
22 (a) B. E. Bursten, J. Am. Chem. Soc., 1982, 104, 1299; (b) B. E.
Bursten and M. R. Green, Prog. Inorg. Chem., 1988, 36, 393.
23 A. J. L. Pombeiro, Eur. J. Inorg. Chem., 2007, 1473.
24 A. B. P. Lever, Inorg. Chem., 1990, 29, 1271.
25 The potential of Mn(ToP)(CO)3 was not measured.
26 S. Bolano, J. Bravo, J. Castro, M. M. Rodrıguez-Rocha,
´
M. F. C. Guedes da Silva, A. J. L. Pombeiro, L. Gonalvi and
M. Peruzzini, Eur. J. Inorg. Chem., 2007, 5523.
´
27 F. Barriere and W. E. Geiger, Acc. Chem. Res., 2010, 43, 1030.
28 In ranking the Re complexes, we note that the cation radicals of the
ReCp and ReCp* complexes are long-lived but in equilibrium with
the equivalent dimer dications (see ref. 15).
29 A. Nafady, P. J. Costa, M. J. Calhorda and W. E. Geiger, J. Am.
Chem. Soc., 2006, 128, 16587.
4 D. D. LeCloux and W. B. Tolman, J. Am. Chem. Soc., 1993,
115, 1153.
5 D. M. Tellers, S. J. Skoog, R. G. Bergman, B. T. Gunnoe and
D. W. Harman, Organometallics, 2000, 19, 2428.
´
6 F. Barriere and W. E. Geiger, Acc. Chem. Res., 2010, 43, 1030.
7 Mn complex: D. R. Laws, D. Chong, K. Nash, A. L. Rheingold
and W. E. Geiger, J. Am. Chem. Soc., 2008, 130, 9859.
8 B. Kovac, L. Klasinc, Z. Raza and V. Sunjic, J. Chem. Soc., Perkin
Trans. 2, 1999, 2455.
9 A. M. Tondreau, J. M. Darmon, B. M. Wile, S. K. Floyd,
E. Lobkobsky and P. J. Chirik, Organometallics, 2009, 28, 3928.
30 S. P. Schmidt, W. C. Trogler and F. Basolo, Inorg. Synth., 1985, 23, 41.
31 (a) S. J. Trofimenko, J. Am. Chem. Soc., 1967, 89, 3170;
(b) S. Trofimenko, J. Am. Chem. Soc., 1967, 89, 6288.
32 J. E. Joachim, C. Apostolidis, B. Kanellakopulos, R. Maier,
N. Marques, D. Meyer, J. Mueller, A. Pires de Matos and
B. Nuber, J. Organomet. Chem., 1993, 448, 119.
33 R. J. LeSuer, C. Buttolph and W. E. Geiger, Anal. Chem., 2004,
76, 6395.
34 (a) G. Gritzner and J. Kuta, Pure Appl. Chem., 1984, 56, 461;
(b) N. G. Connelly and W. E. Geiger, Chem. Rev., 1996, 96, 877.
35 M. J. Shaw and W. E. Geiger, Organometallics, 1996, 15, 13.
c
2178 New J. Chem., 2011, 35, 2169–2178
This journal is The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2011