Para-Substituted diphenylborylated organocobaloximes: Effects of substituents on conformation and redox properties

Article abstract of DOI:10.1016/S0022-328X(00)00043-7

Some new derivatives of organocobaloximes containing para-substituted diphenylboryl groups, RCo(DH)_2-n(DB(p-XPh)₂)_nL (R=alkyl or aryl group, L=N-MeIm, Py or H₂O, X=OCH₃, CH₃ or Cl, n=1 or 2) have been synthesized. The X-ray structures and the ¹H-NMR spectra are compared with those of the corresponding RCo(DH)_2-n(DBPh₂)_nL and RCo(DBF₂)₂L complexes. The insertion of X groups in the phenyl rings does not significantly affect the equatorial Co-N distances, whereas the Co-Py distances increase slightly in the order (DB(p-OCH₃Ph)₂)₂<(DB(p-ClPh) ₂)₂<(DBF₂)₂. ¹H-NMR spectra suggest that the conformational distribution in solution is similar to that observed in the corresponding BPh₂ derivatives. Electrochemical studies on the corresponding MeCo(DB(p-XPh)₂)₂H₂O compounds show a mono-electron Co(III)/Co(II) transfer reaction followed by two parallel reactions: (a) mono-electron Co(II)/Co(I) transfer; (b) homolytic dissociation of the Co-C bond with the formation of Co(I) species, the relative rates of the two processes being dependent on X. As the electron-withdrawing power of the equatorial ligand increases, the reduction potentials associated with both Co(III)/Co(II) and Co(II)/Co(I) processes shift towards more positive values, indicating a decrease of electron density on the Co atom. The effects are comparable with those observed by changing the axial ligands.

Full text of DOI:10.1016/S0022-328X(00)00043-7

www.elsevier.nl/locate/jorganchem
Journal of Organometallic Chemistry 601 (2000) 114–125
para-Substituted diphenylborylated organocobaloximes: effects of
substituents on conformation and redox properties
Fioretta Asaro, Renata Dreos *1, Giorgio Nardin, Giorgio Pellizer, Silvia Peressini,
Lucio Randaccio *2, Patrizia Siega, Giovanni Tauzher, Claudio Tavagnacco
Dipartimento di Scienze Chimiche, Uni6ersita` Degli Studi di Trieste, 6ia Licio Giorgieri 1, I-34127 Trieste, Italy
Received 8 October 1999; received in revised form 24 November 1999; accepted 5 January 2000
Abstract
Some new derivatives of organocobaloximes containing para-substituted diphenylboryl groups, RCo(DH)2−n(DB(p-XPh)2)nL
(R=alkyl or aryl group, L=N-MeIm, Py or H2O, X=OCH3, CH3 or Cl, n=1 or 2) have been synthesized. The X-ray
1
structures and the H-NMR spectra are compared with those of the corresponding RCo(DH)2−n(DBPh2)nL and RCo(DBF2)2L
complexes. The insertion of X groups in the phenyl rings does not significantly affect the equatorial CoꢀN distances, whereas the
1
CoꢀPy distances increase slightly in the order (DB(p-OCH3Ph)2)2B(DB(p-ClPh)2)2B(DBF2)2. H-NMR spectra suggest that the
conformational distribution in solution is similar to that observed in the corresponding BPh2 derivatives. Electrochemical studies
on the corresponding MeCo(DB(p-XPh)2)2H2O compounds show a mono-electron Co(III)/Co(II) transfer reaction followed by
two parallel reactions: (a) mono-electron Co(II)/Co(I) transfer; (b) homolytic dissociation of the CoꢀC bond with the formation
of Co(I) species, the relative rates of the two processes being dependent on X. As the electron-withdrawing power of the equatorial
ligand increases, the reduction potentials associated with both Co(III)/Co(II) and Co(II)/Co(I) processes shift towards more
positive values, indicating a decrease of electron density on the Co atom. The effects are comparable with those observed by
changing the axial ligands. © 2000 Elsevier Science S.A. All rights reserved.
Keywords: Organocobaloximes; p-Substituted diphenylborinic groups; Conformational equilibria; X-ray structures; Electrochemical properties
1. Introduction
from ‘up’ to ‘down’ for the monoborylated derivatives
(Scheme 1), and from ‘up–down’ to ‘down–down’ for
the diborylated ones with increasing steric bulk of R
(Scheme 2) [4]. Pure ‘up–up’ forms have not been
evidenced yet either in solution or in solid state, if
the pentacoordinate Co(I) species, PyCo(DBF2)−2 , is
excluded [6].
X-ray diffraction studies show that when the complex
assumes a conformation in which at least one phenyl of
the BPh2 group faces a planar neutral ligand L, the
latter is forced into an orientation that bisects the
five-membered rings of the equatorial moiety. This
orientation is quite unusual in cobaloximes, but is
generally observed in iminocobaloximes, {RCo[(DO)-
(DOH)pn]L}+ cations, and leads to a lengthening of
the axial CoꢀN [7,8]. Ligands with good p-acceptor
properties stabilize ‘down–down’ conformation [9], so
confirming the role of p–p interactions between the
phenyls of the BPh2 bridge and the axial ligands, as
previously found for LFe(II)(DBPh2)2L% systems [5].
Bis(dimethylglyoximate) complexes in which the pro-
tons of the oxime groups are formally replaced by boryl
bridges are well known [1]. In organocobaloximes,
RCo(III)(DH)2L, the replacement of the O···H···O
bridge with BF2, RCo(DBF2)2L, affects both the ge-
ometry of the complexes [2] and the electron density at
the central metal ion [3]. Organocobaloximes contain-
ing BPh2 bridges, RCo(DH)2−n(DBPh2)nL (R=alkyl
or aryl group, L=neutral ligand, n=1 or 2), may
adopt different conformations in solution and 1H-
NMR spectroscopy allows the conformational equi-
librium to be studied by exploiting the anisotropy
effects both of the BPh2 phenyls and of the axial
ligands [4,5]. The conformational equilibrium shifts
1*Corresponding author. Tel.: +39-40-6763902; fax: +39-40-
6763903; e-mail: dreos@univ.trieste.it
2*Corresponding author. Tel.: +39-40-6763902; fax: +39-40-
6763903.
0022-328X/00/$ - see front matter © 2000 Elsevier Science S.A. All rights reserved.
PII: S0022-328X( 00)00043-7

F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
115
In order to investigate the effects of a para substitu-
tion in the phenyls on conformation and electron
affinity of the redox center, we report here the synthesis
and the characterization of several derivatives of
tives. The completeness of the reaction was checked by
1H-NMR spectroscopy. The complexes were identified
1
by H- and 13C-NMR spectroscopy and, in some cases,
by elemental analysis.
organocobaloximes
containing
para-substituted
diphenylboryl groups.
2.1.1. RCo(DH)(DB(p-XPh)2)L deri6ati6es
(L=N-MeIm or Py)
A total of 0.1 g RCo(DH)2L was dissolved in about
50 ml of CH2Cl2 and an equimolar amount of borinic
acid was added, together with a small excess of L in
order to avoid the dissociation of the axial base. The
solution was heated at 35°C for times varying from 1 to
3 days. Partial evaporation of the solvent and the
addition of 2–3 ml of i-PrOH afforded yellow powders,
which were recrystallized from CH2Cl2/i-PrOH.
2. Experimental
RCo(DH)2L [10], RCo(DBF2)2H2O [10] and
RCo(DBPh2)2L [4] were synthesized as previously
described.
The (p-ClPh)2BOH, (p-OCH3Ph)2BOH and (p-
CH3Ph)2BOH acids were prepared according to the
procedure of Povlock and Lippincott [11]. Solvent and
other reagents have been commercially purchased.
2.1.2. RCo(DB(p-XPh)2)2L deri6ati6es (L=N-MeIm
or Py)
2.1. Syntheses
A total of 0.1 g of the corresponding RCo(DH)2L
was dissolved in about 50 ml of CH2Cl2 with an
eight-fold excess of the borinic acid. Some drops of L
were added in order to avoid the dissociation of the
axial base. The solution was heated at 35°C for times
varying from 1 to 3 days. Partial evaporation of the
solvent and the addition of 2–3 ml of i-PrOH afforded
yellow powders, which were recrystallized from
CH2Cl2/i-PrOH.
We report the general procedure for the synthesis of
the complexes RCo(DH)2−n(DB(p-XPh)2)nL (R=
alkyl or aryl group, L=N-MeIm, Py, or H2O, X=
OCH3, CH3 or Cl, n=1 or 2). The reaction time
depends on both the bulk of the axial ligands [4] and
the nature of the substituent X in para position in the
phenyls, being considerably longer for the p-Cl deriva-
2.1.3. RCo(DB(p-XPh)2)2H2O deri6ati6es
A total of 0.1 g of the corresponding RCo(DH)2H2O
was suspended in about 50 ml of CH2Cl2 saturated with
water; then acetone was added until dissolution.
Borinic acid in molar ratio 3:1 with complex was added
to the solution, which was heated at 40°C for 1 or 2
days. Partial evaporation of the solvent afforded the
products. All the complexes were recrystallized from
CH2Cl2/i-PrOH.
Scheme 1.
2.1.4. MeCo(DB(p-OCH3Ph)2)2H2O
A total of 0.1 g MeCo(DBp-OCH3Ph2)2N-MeIm was
dissolved in 40 ml of dimethylsulfoxide. The solution
was acidified with HClO4 (pH about 3) and left for
about 10 min. Addition of water caused the precipita-
tion of MeCo(DB(p-OCH3Ph)2)2H2O, sparingly soluble
in water.
2.2. NMR measurements
1H- and 13C-NMR spectra were recorded on a JEOL
EX-400 (1H at 400 MHz and 13C at 100.4 MHz) from
CDCl3 solutions with TMS as internal standard.
2.3. X-ray structure determinations
Crystallographic data for MeCo(DH)(DBPh2)Py (1),
MeCo(DBPh2)2Py (2), MeCo(DB(p-ClPh)2)2Py (3), and
Scheme 2.

116
F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
Table 1
Crystal data and experimental conditions of complexes 1, 2, 3, and 4
1
2
3
4
Empirical formula
Formula weight
C26H31BCoN5O4
547.3
C
711.3
38H40B2CoN5O4
C38H36B2Cl4CoN5O4
849.1
C42H48B2CoN5O8 · CH2Cl2
916.3
Crystal dimensions (mm)
Unit cell dimensions
0.2×0.5×0.2
0.2×0.2×0.7
0.2×0.3×0.2
0.2×0.4×0.7
,
a (A)
8.651(2)
16.275(7)
18.727(4)
90
93.57
90
8.353(2)
10.479(3)
11.627(3)
66.82(3)
73.27(2)
72.12(2)
874.0(4)
1
16.740(3)
10.262(2)
23.459(3)
90
99.41(10)
90
9.305(2)
15.930(4)
16.607(4)
70.38(2)
73.44(2)
83.57(2)
2222.1(9)
2
,
b (A)
,
c (A)
h (°)
i (°)
k (°)
V (A )
3
,
2633(1)
4
3976(1)
4
Z
(
(
Space group
P21/n
„-scan
3831
P1
P21/n
„-scan
2489
P1
Absorption correction
Independent reflections
Parameters refined
R1
wR2
Goodness-of-fit
„-scan
3816
252
„-scan
4215
559
159
247
0.093
0.183
0.94
0.071
0.172
1.15
0.106
0.184
0.92
0.087
0.213
1.12
Table 2
Comparative geometrical values for MeCo(DH)2−n(DB(p-XPh)2)n Py
X
n
1
1
Complex
Ref. [4b]
1
CoꢀNeq(mean)
CoꢀC
CoꢀNax
2.014(9)
2.083(9)
O···O
d(Co)
0.009
0.033
H(N-MeIm)
1.869(8) a
1.878(8) b
1.876(9) a
1.869(10) b
1.867(4)
1.995(12)
2.000(11)
2.469 a
2.510 b
H
2.438 a
2.534 b
H
Cl
OCH3
2
2
2
2
3
4
2.098(5) c
1.998(9)
2.032(6)
2.098(5) c
2.106(8)
2.086(6)
2.526
2.542 2.528
2.524 2.519
0
1.864(8)
1.867(5)
0.076
0.082
a On the side of BPh2 group.
b On the side of the oxime bridge.
c The values are a consequence of the disorder due to the location of Co on the symmetry center, so that they cannot be discussed.
2.4. Electrochemical measurements
MeCo(DB(p-OCH3Ph)2)2Py (4) are collected in Table 1.
Structural data of interest are given in Table 2.
All the electrochemical measurements were made in
dimethylsulfoxide (DMSO) solution that was prepared
by holding the commercial product over NaOH for 3 h
at 90°C and distilling it at reduced pressure [13]. Te-
traethylammonium perchlorate (TEAP) vacuum dried
at 30°C was the supporting electrolyte. (CAUTION:
perchlorate salts are potentially explosive and should be
handled with great care.)
Polarographic and cyclic voltammetry (CV) measure-
ments were performed using an Amel 552 Potentiostat/
Galvanostat connected with an Amel 568 function
generator. For fast CV a home-made potentiostat [14]
equipped with positive feedback and driven by a Hew-
lett–Packard function generator 3314A was used. The
measurements were always made at 2590.1°C under
Ar or N2 atmosphere in a three-electrode thermostated
Crystals of complexes 1–4, suitable for X-ray struc-
ture analysis, were obtained by slow diffusion of
CH2Cl2 solutions in i-PrOH; however, in the case of
complex 1 the size of the crystals was very small.
Preliminary examination and data collection were per-
,
formed with Mo–Ka radiation (u=0.70930 A) on an
Enraf–Nonius CAD4 diffractometer with a graphite
monochromator. Empirical absorption correction,
based on „-scan data, was applied. The structures were
solved by conventional Patterson, Fourier analysis and
refined by the least-square methods. H atoms were
located at calculated positions. Final positional
parameters and B values are reported as supplementary
material. All calculations were performed using SHELXS
and SHELXL programs [12].

F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
117
jacket cell. A dropping mercury electrode (DME) was
used as the working electrode for polarography, while a
Metrohom 663 VA Stand for Hg was used for CV. The
reference was a saturated aqueous NaCl ꢀ calomel elec-
trode (SCE), contained in a glass tube separated from
the solution by a glass frit of medium porosity which
was located very close to the tip of the working elec-
trode to minimize the ohmic drop, and filled with the
supporting electrolyte solution. The counter electrode
was a Pt ring or rod directly dipped in the solution. The
peak potentials and currents were evaluated after sub-
traction of the background current.
in the analogous MeCo(DH)(DBPh2)N-MeIm complex
[4]. In complex 2 the cobalt atom lies on a crystallo-
graphic symmetry center so that the equatorial moiety
shows an ‘up–down’ conformation and the axial lig-
ands are superimposed. The pyridine ligand is rotated
by about 25° with respect to the ideal face-to-face
contact with the axial phenyl of the BPh2 unit (orienta-
tion B).
The torsion angles l=N2%–Co–N5–C17 and l%=
N1–Co–N5–C17 of −17 and 65°, respectively, indi-
cate the relative position of the pyridine plane with
respect to the equatorial moiety.
Controlled potential reduction (CPR) was made on a
stirred Hg pool of about 12 cm2 surface area: in that
case the counter electrode was separated from the solu-
tion by a glass frit filled with the supporting electrolyte
solution. The charge passed was determined by an
Amel 731 integrator.
Complexes 3 and 4 show a very similar ‘up–down’
extended chair conformation. The equatorial moieties
of 3 and 4 have an ca. C2h symmetry, if the orientation
of the equatorial p-OCH3Ph rings at boron is not
considered (Fig. 2). The pyridine, in orientation B,
faces the axial p-ClPh ring in 3 and the analogous
p-OCH3Ph ring in complex 4. The corresponding l and
l% values are −33.8 and 47.8 for 3 and 34.8 and
−47.2° for 4, respectively. Consequently, the plane of
the pyridine ring in 3 and 4 is nearly perpendicular to
the plane defined by the cobalt and two boron atoms.
Comparison of the O···O distances in 1 indicates a
3. Results
3.1. Syntheses
The reactivity of organocobaloximes with para-sub-
stituted diphenylborinic acids strictly resembles that
with diphenyl borinic anhydride [4]. When L is a nitro-
gen base, generally, either mono or diphenylborilated
derivatives could be isolated, depending on the acid/
complex ratio. In some cases, for bulky axial ligands
such as phenyl or trans-b styryl, the diborylated com-
plex could not be isolated and NMR spectra always
revealed mixtures of mono and diborylated complexes.
When L was H2O only diphenylborylated complexes
were recovered. The aquaderivatives were generally
obtained by heating the corresponding organoaqua-
cobaloxime in presence of borinic acid, but for
MeCo(DB(p-OCH3Ph)2)2H2O this method was unsuc-
cessful, the reaction product being a complex contain-
ing one boryl bridge and one molecule of
(p-OCH3Ph)2BOH coordinated trans to methyl. Similar
complexes have been evidenced previously, when
organocobaloximes were treated with diphenylborinic
anhydride in anhydrous solvent [15]. Authentic
MeCo(DB(p-OCH3Ph)2)2H2O could be recovered by
hydrolysis of MeCo(DB(p-OCH3Ph)2)2N-MeIm in
DMSO and successive addition of water to the reaction
mixture.
,
shortening of about 0.1 A on the side of the oxime
bridge (Table 2). In 2, 3 and 4 all the O···O distances
,
are very similar, being in the range 2.519–2.542 A. The
presence of one or two boron bridges or the insertion of
substituents in the phenyl rings does not seem to affect
significantly the equatorial CoꢀN bond distances (Table
2). Also the CoꢀCH3 bond distances do not appear to
be influenced by the nature of the equatorial ligand, the
value of 2.098(5) for 2 being an artefact due to the axial
CH3/Py disorder imposed by the symmetry center. Only
in 4 does the CoꢀC of 2.032(6) appear slightly but
significantly longer than the values found in 1 and 3
(Table 2). The neutral ligand in 1 and in the analogous
N-MeIm derivative [4] has orientation close to A, the
CoꢀPy distances being 0.07 A longer, due to the larger
bulk of Py. This also affects the Co displacement out of
the equatorial 4-N donors towards L, which is 0.033 A
in complex 1 and 0.009 A in the N-MeIm analogue. It
has been observed in cobaloximes and iminocoba-
loximes [7] that the CoꢀNaxial distances are appreciably
affected by the orientation of the planar L ligand, being
shorter in orientation A, as occurs in N-MeIm deriva-
tives with n=1 (2.014(9) A), than in orientation B, as
occurs in N-MeIm derivatives with n=2 [4], (2.068(7)
,
,
,
,
,
A). Comparison of the analogous Py derivatives with
3.2. Structural results
n=1 and 2, respectively (Table 2) is hampered by the
disorder in 2. The Co displacement out of the 4-N
equatorial plane towards N pyridine atom varies from
The ORTEP drawings of the complexes 1 and 2 are
depicted in Fig. 1 and those of 3 and 4 in Fig. 2. In
complex 1, the axial CH3 group faces the axial phenyl
of the BPh2 unit and the pyridine nearly bisects the
six-membered ring (orientation A), as previously found
,
,
0.03 A in complex 1 to 0.08 A in 3 and 4. Only in 2,
owing to the special position of the metal, the cobalt
atom lies in the coordination basal plane.

118
F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
Fig. 1. The ORTEP drawing and the atom numbering scheme of the complexes MeCo(DH)(DBPh2)Py (1) and MeCo(DBPh2)2Py (2) (thermal
ellipsoids at 30% probability).

F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
119
Fig. 2. The ORTEP drawing and the atom numbering scheme of the complexes MeCo(DB(p-ClPh)2)2Py (3) and MeCo(DB(p-OCH3Ph)2)2Py (4)
(thermal ellipsoids at 30% probability).

Table 3
1H-NMR chemical shift of RCo(DH)n(DB(p-XBPh)2)2−nN-MeIm complexes a,b
Complexes
R
CH3 eq.
N-MeIm
H-2
B(p-XPh)2
a
b
g
d
H-5, H-4
CH3
ortho
meta
OCH3
MeCo(DH)2N-MeIm c
MeCo(DH)(DB(p-OCH3Ph)2)N-MeIm
0.72
0.22
2.13
2.15–2.36
7.44
7.40
6.78, 6.94
6.72, 7.01
3.66
3.59
7.18
7.47
7.09
6.63
6.74
6.79
7.20
6.74
3.74–3.76
3.72–3.74
MeCo(DB(p-OCH3Ph)2)2N-MeIm
0.42
2.41
6.02
6.46, 6.48
3.30
n-PrCo(DH)2N-MeIm c
n-PrCo(DH)(DB(p-OCH3Ph)2N-MeIm
1.52
1.26
0.94
0.61
0.78
0.44
2.13
2.19–2.38
7.42
6.90
6.76, 6.95
6.60, 6.73
3.62
3.48
7.29
7.29
6.98
7.24
6.69
6.76
6.55
6.75
/
3.74–3.75
3.69–3.73
n-PrCo(D(Bp-OCH3Ph)2)2N-MeIm
1.51
0.44
0.29
2.43
5.94
6.21, 6.24
3.24
n-BuCo(DH)2N-MeIm c
n-BuCo(DH)(DB(p-OCH3Ph)2)N-MeIm
1.52
1.28
0.87
0.53
1.18
0.81
0.78
0.64
2.12
2.18–2.38
7.42
6.93
6.75, 6.94
6.61, 6.75
3.62
3.48
7.28
7.29
6.98
7.24
6.69
6.75
6.55
6.74
3.74
n-BuCo(DB(p-OCH3Ph)2)2N-MeIm
1.52
0.64
2.42
5.97
6.23, 6.25
3.25
3.70–3.73
PhCo(DH)2N-MeIm c
PhCo(DH)(DB(p-OCH3Ph)2)2N-MeIm
7.39
7.54
6.89–6.94 (m+p)
6.97 (m+p)
2.04
2.18–2.34
7.57
5.88
6.79, 7.08
6.03, 6.35
3.66
3.29
6.89
7.54
6.73
7.41
6.46
6.86
6.34
6.80
3.69–3.79
3.66–3.77
PhCo(D(Bp-OCH3Ph)2)2N-MeIm
7.73 (o)
2.40
5.50
5.47, 5.33
6.08, 6.37
3.01
MeCo(DH)(DB(p-ClPh)2)N-MeIm
MeCo(DB(p-ClPh)2)2N-MeIm
0.22
0.53
2.16–2.38
2.47
6.72
5.91
3.59
3.36
6.99
7.20
6.94
7.14
601
(
PhCo(DH)(DB(p-ClPh)2)N-MeIm
PhCo(D(Bp-ClPh)2)2N-MeIm
7.45 (o)
7.55 (o)
6.92 (m+p)
6.94 (m+p)
2.20–2.36
2.43
6.40
5.53
5.93, 5.97
5.31, 5.66
3.39
3.15
6.89
7.53
6.75
7.41
7.11
7.17
6.87
7.52
6.71
7.40
6.83
7.24
6.75
7.20
6.87
6.97
6.67
7.09
6.56
7.05
1
125
MeCo(DB(p-CH3Ph)2)2N-MeIm
PhCo(DH)(DB(p-CH3Ph)2)2N-MeIm
PhCo(D(Bp-CH3Ph)2)2N-MeIm
0.39
2.40
6.15
5.95
5.52
6.41, 6.51
6.00, 6.30
5.35, 5.29
3.27
2.95
2.95
2.24–2.20
2.19–2.30
2.14–2.20
7.52 (o)
7.73 (o)
2.18–2.35
2.41
a l in ppm from TMS, CDCl3 solutions.
b Some of the monoborylated complexes were not isolated but directly prepared in the NMR tube.
c From Ref. [4].

F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
121
1
3.3. H-NMR studies
bridge protons. All the compounds studied in this paper
show the signals of two different phenyls, even those
containing two boryl bridge. The spectral pattern of
each phenyl reflects its fast exchange between the axial
and the equatorial situations, the weights of which
depend on the conformational equilibrium [4,5]. The
proton spectra of B(p-XPh)2 groups correspond to
AA%BB% spin systems but the extreme case of an A4 spin
system occurs for the more shielded phenyl of
PhCo(DB(p-ClPh)2)2N-MeIm.
The shielding of the phenyl protons, beside on X,
depends on the through-space effects of the axial lig-
ands, and therefore on the conformation. This was
exploited in order to determine the influence of R on
the preferred conformation in solution. We have taken
the difference between the chemical shifts of the pro-
tons of each X-substituted compound and those of the
corresponding protons in the parent compounds (X=
H). The values so obtained for a given position (ortho
or meta to boron), and a given X are independent of R
and of the number of boryl bridges (Table 4) and close
to the values of substituent induced shifts reported in
literature [16]. This implies that the through-space ef-
fects caused by the axial ligands are the same both in
the compounds with X=CH3, OCH3 and Cl and in
that with X=H, and confirms that the conformational
equilibrium is substantially independent of X. Accord-
ingly, the dependence on X of the shielding of N-MeIm
methyl and of the g protons of the n-Pr and n-Bu axial
ligands (see above) reflects changes of magnetic an-
isotropy and electric dipoles due to substitution of the
BPh2 phenyls, rather than modifications of the confor-
mational equilibrium.
3.3.1. Axial ligands
Only one set of signals is observed for each axial
ligand because the interconversion among conformers is
fast on the NMR time-scale in CDCl3 solution at room
temperature. The proton chemical shift of the methyl of
N-MeIm has proved to be a valid parameter for investi-
gating the conformational equilibria, because it re-
sponds almost exclusively to the through-space effects
of the BPh2 aromatic rings [4]. In the MeCo(DH)(DB-
(p-XPh)2)N-MeIm compounds (X=OCH3 and Cl) its
resonance frequency does not depend significantly on X
(Table 3). This indicates that the prevailing conforma-
tion is ‘up’, as already found for the parent compound
(X=H) [4]. In the PhCo(DH)(DB(p-XPh)2)N-MeIm
complexes the methyl of N-MeIm is much more
shielded, indicating that the ‘down’ conformation is
preferred, like when X=H. In this case the magnitude
of the shielding shows a clear dependence on X, de-
creasing in the order HꢀCH3\OCH3\Cl.
In the RCo(DB(p-XPh)2)2N-MeIm compounds the
methyl of N-MeIm is more shielded that in the corre-
sponding monoborylated species and the main confor-
mation is ‘up–down’ for R=Me and ‘down–down’ for
R=Ph, as for the derivatives with unsubstituted BPh2
bridges. The effect decreases on changing X in the
above reported order.
In the n-Pr and n-Bu derivatives, the R protons
bonded to the g carbon are affected only by B(p-XPh)2
through-space effects [4]. The RCo(DH)2−n(DB(p-
OCH3Ph)2)nN-MeIm (R=n-Pr and n-Bu) compounds
behave like the corresponding ones with unsubstituted
BPh2 groups, except for the slightly smaller shielding
both on the methyl of N-MeIm and on the protons at
the g carbons.
3.4. Electrochemical studies
The electrochemical studies were performed on a
series of aquaderivatives: MeCo(DBF2)2H2O (5), MeCo-
(DB(p-ClPh)2)2H2O (6), MeCo(DB(p-OCH3Ph)2)2H2O
(7), MeCo(DBPh2)2H2O (8), MeCo(DB(p-CH3Ph)2)2-
H2O (9) and MeCo(DH)2H2O (10).
All the measurements were carried out in DMSO, as
this was the only solvent in which all the above com-
plexes were sufficiently soluble.
3.3.2. Equatorial ligand
The equatorial methyls in the RCo(DH)2−n(DB(p-
XPh)2)nN-MeIm resonate close to those of the corre-
sponding complexes with X=H. Further insight is
gained through examination of the spectra of the boryl
Table 4
X-substituent-induced shifts on the phenyl protons in RCo(DH)2−
n(DB(p-XPh)2)nN-MeIm compounds
3.4.1. Polarography
The polarography in DMSO and 0.1 M TEAP in the
range from +0.3 to −2.0 V shows two distinct mono-
electronic reduction waves that are associated with the
Co(III)/Co(II) (wave I) and Co(II)/Co(I) (wave II) pro-
cesses, respectively. Some polarographic characteristic
data are summarized in Table 53. The height of the
X
Dl (ppm) a
Meta
Ortho
CH3
OCH3
Cl
−0.20
−0.42
−0.04
−0.15
−0.12
−0.11
3 Serious problems were sometimes encountered upon electrochem-
ical reduction due to adsorption signals that limit the possibility
of getting high-precision results, especially with MeCo(DB(p-
ClPh)2)2H2O and MeCo(DBPh2)2H2O.
a Dl is the difference between the chemical shifts of the protons of
the X-substituted compound and those of the corresponding protons
in the parent compounds (X=H).

122
F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
waves is linearly related to the concentration of the com-
plexes up to 2 mM.
From the analysis of the plot of log [i/(il−i )] versus
E, wave I is associated with an electrochemically quasi-
reversible or irreversible process [17], while wave II is as-
sociated with a reversible process for 6 and 9 and
quasi-reversible one for the others. Both processes are
monoelectronic, as it was proved by the CPR experi-
ments made on the plateau of the first wave (see below),
but for 7, 8 and 9 the ratio il(I)/il(II) is about 1 (Fig. 3
(A)), whereas for 5, 6 and 10 it is greater than 1 (Fig.
3(B)).
The E1/2 values relative to the Co(III)/Co(II) electron
transfer become 530 mV more negative on going from 5
to 10; the E1/2 values relative to the Co(II)/Co(I) electron
transfer follow a more complicated pattern (Fig. 4).
3.4.2. Controlled potential reduction
After exhaustive CPR corresponding to 1 faraday
mol−1 at the potential of the plateau of the first wave,
made under inert gas atmosphere, both the cathodic
waves disappear while anodic signals develop: the height
of those waves is linearly proportional to the number of
Coulombs passed.
The solution from a typical yellow–orange color be-
comes green–blue and is stable for several minutes under
nitrogen atmosphere at ambient temperature. The polar-
ograpy made after the CPR shows the development
of new cathodic and anodic waves due to the reduction
and oxidation of the products resulting after reduction.
3.4.3. Cyclic 6oltammetry
The CV results on an Hg drop electrode are in agree-
ment with the polarographic data and are summarized in
Table 5. All the examined compounds show a couple
of cathodic peaks and each of them is coupled with an
anodic partner, at least at sufficiently high scan rate.
The first cathodic/anodic peak couple is associated
with a mono-electron Co(III)/Co(II) transfer while the
second cathodic/anodic peak couple proves another
mono-electron transfer that is associated with the
Co(II)/Co(I) process at more negative potential (Fig. 3).
For 7, 8 and 9 CV made at low scan rate (0.1 V s−1
)
shows that the ratios between the cathodic and anodic
current peaks, Ipc(I)/Ipa(I) and Ipc(I)/Ipc(II), are about
1 (Fig. 3 (C)), while in the case of 5, 6 and 10 the
ratios Ipc(I)/Ipa(I) and Ipc(I)/Ipc(II) are greater than 1
at low scan rates (Fig. 3(D)) and approach 1 as the scan
rate increases; meanwhile the Epc(I) shifts towards
more positive values [18].
4. Discussion
Present results confirm previous findings [4,5] that in
Co(DBPh2)2 complexes, the nature of the axial ligands

F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
123
tion is found in the solid state, while the ‘down–down’
conformation, with both the axial phenyl groups facing
the acetonitrile ligand, is observed when R=n-Pr and
b-styryl. A similar ‘down–down’ conformation, due to
a large difference in bulk between the axial ligands, has
also been reported for the octahedral PyFe(II)-
(DBPh2)2PPh2Me complex [5f]. The only case in which
an ‘up–up’ conformation is observed is the pentacoor-
dinate Co(I) species, PyCo(DBF2)−2 [6], where, because
of the lack of the trans axial ligand, Py can assume the
preferred orientation A, forcing the equatorial moiety
into that conformation. Similarly, in MeCo(DBPh2)-
(DH)N-MeIm and in MeCo(DBPh2)(DH)Py the planar
ligand assumes the preferred orientation A, forcing the
axial phenyl of the BPh2 bridge to face the alkyl group.
It is of interest to compare the coordination geometry
of the MeCo(chel)Py, when chel is varied. The mean
values of the CoꢀN equatorial distances are very similar
in the boryl derivatives and shorter than in cobaloximes
(Table 6), possibly due to the strong decrease of elec-
tron density on the metal center on substitution of the
oxime bridges with boryl bridges (see below). Data of
Table 6 suggest that the CoꢀPy distance is longer in
orientation B, but for the same orientation it appears to
increase slightly in the order (DB(p-OCH3Ph)2)2B
(DB(p-ClPh)2)2B(DBF2)2. This might be indicative of
some electronic cis influence. The CoꢀCH3 distances are
very similar in all but one of the complexes of Table 6.
Fig. 3. Polarography in DMSO+0.1 M TEAP at 25°C of 8 (A) and
5 (B). Cyclic voltammetry in the same solvent of 8 (C) and 10 (D) at
a scan rate of 0.1 V s−1 on Hg electrode. The concentration of the
complexes is about 1 mM.
,
The value of 2.032(6) A in the (DBp-OCH3Ph2)2 deriva-
tive is slightly greater than those in the other ones. If
significant, such a difference could be attributed to the
effect of the para methoxy substituent; unfortunately,
the corresponding distances in the parent MeCo-
(DBPh2)2Py are not available for a better comparison
(see above).
Fig. 4. Schematic representation of the relative stability of the oxida-
tion states Co(III)/Co(II)/Co(I) for the complexes 5–10 in DMSO+
0.1 M TEAP at 25°C.
determines the conformation of the equatorial moiety,
through their electronic and steric interactions with the
BPh2 groups. When both axial ligands have similar and
moderate bulk and display similar electronic interac-
tions with phenyl group, the ‘up–down’ conformation
is the preferred one in the solid state. This implies that
planar axial ligands such as N-MeIm or Py assume
orientation B. This observation suggests that the
‘down–down’ conformation, which leads to the ‘sand-
wiching’ of one axial ligand by the phenyl groups,
should be less stable than the ‘up–down’ conformation
unless the latter ligand is involved in strong p–p attrac-
tive interactions with the phenyl rings. In this case, the
‘down–down’ conformation is stabilised by the ‘sand-
wiching’ electronic interactions, as found in the
1
The H-NMR spectra suggest that also the distribu-
tion of conformations in solution for the RCo(DH)2−
n(DB(p-XPh)2)nN-MeIm compounds and that observed
in the corresponding BPh2 derivatives are the same,
being mainly ‘up’ and ‘up–down’ for R=alkyl and
‘down’ and ‘down–down’ for R=Ph, respectively. On
one hand, the independence of the conformational
equilibrium from the phenyl substituents candidates
these molecules as suitable systems for studying varia-
tions of through-space effects induced by substitution
of aromatics and related changes in magnetic an-
isotropy and bond electric dipoles [19]. On the other
hand, it induces a reconsideration of the relative effec-
tiveness of the steric and p–p interactions in determin-
ing the conformation of these complexes. Indeed,
according to the model of Hunter and Sanders the
major contributions to the p–p interactions come from
the electrostatic and van der Waals interactions [20]. At
MeCo(DBPh2)2TCNE
(TCNE=tetracyanoethylene)
[9]. On the contrary, when the difference in bulk be-
tween the two axial ligand increases, the ‘up–down’
conformation becomes less stable. This is well illus-
trated by the structures of
a
series of the
,
RCo(DBPh2)2(NCCH3) complexes, with R having dif-
ferent bulk [9]. For R=Me the ‘up–down’ conforma-
interplanar separations greater than 3.4 A the latter is
always attractive and is roughly proportional to the

124
F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
Table 6
Comparison of coordination geometry for MeCo(chel)Py complexes with several equatorial chel ligands
,
,
,
Complexes
CoꢀNeq (mean, A)
CoꢀMe (A)
CoꢀPy (A)
Orientation
MeCo(DBF2)2Py a
1.861(13)
1.867(5)
1.864(8)
1.873(10)
1.899(8)
2.007(8)
2.032(6)
1.998(9)
2.000(11)
1.998(6)
2.119(4)
2.086(6)
2.106(8)
2.083(9)
2.068(3)
B
B
B
A
A
MeCo(DB(p-OCH3Ph)2)2Py
MeCo(DB(p-ClPh)2)2Py
MeCo(DH)(DBPh2)Py
MeCo(DH)2Py b
a From ref. [2].
b From ref. [22].
Scheme 3.
one, therefore comparable to that due to a variation of
the R group from Me to Ph [3]. On the whole, as the
electron-withdrawing power of the equatorial ligand
increases, the reduction potentials associated with both
Co(III)/Co(II) and Co(II)/Co(I) processes shift towards
more positive values, indicating the decrease of electron
density on Co atom (Fig. 4). Noticeably, the coba-
loxime 10 is stable in the Co(III) oxidation states at
potentials at which all the other examined compounds
are in the Co(I) oxidation state. The stability range of
the Co(II) state is about the same for the complexes 6
and 8 but it slightly increases for the complexes 7, 9, 10
and is almost double for the complex 5 (Fig. 4).
area of p–p overlap. The former is very geometry
dependent: it varies from attractive for an edge-on or
an offset p stacked geometry to repulsive for a face-to-
face stacked geometry. The presence of a strongly po-
larizing group should affect the electron density on the
aromatic ring and enhance or relieve the electrostatic
repulsion, so influencing the overall geometry of the
system. This really happens, for example, in the so-
called molecular clips [21]. In the present case the
insertion of substituents in the phenyl ring does not
cause relevant structural modification, suggesting that
more striking variation in the p properties (e.g. the
substitution of TCNE for Py as axial ligand) are re-
quired in order to alter conformation both in solution
and in solid state.
On the contrary, the insertion of a substituent in the
phenyl ring affects remarkably both the electronic den-
sity at the metal center and the rate of the coupled
chemical reaction following the electron transfer. It was
already demonstrated that the substitution of O···H···O
bridges with BF2 groups both in cobaloximes and in
iminocobaloximes strongly decreases the electron den-
sity on the central metal ion [3]. A direct comparison of
these data with the present ones is not possible because
different solvents have been used. However, data in
Table 5 show that, for methylcobaloxime derivatives,
the electrochemical behavior of the complexes 6, 7, 8,
and 9, containing B(p-XPh)2 groups, is intermediate
between that of the unsubstituted cobaloxime 10 and
that of the complex 5 containing BF2. Within the series
6–9, the substituents on the aromatic ring also influ-
ence the electron density on central Co atom, although
to a lesser extent. This effect is of about 100 mV for the
first reduction step and about 160 mV for the second
The increase of the ratio of the CV peak currents
Ipc(II)/Ipc(I) up to 1 on increasing the scan rate and the
shift of the potentials with the scan rate towards more
positive values prove that there is a coupled chemical
reaction following the first electron transfer. Further-
more, the disappearance of the second cathodic peak in
CV when the CPR is made at the potential of the first
one is in agreement with the reaction in Scheme 3 [18].
According to this Scheme the first electron transfer is
followed by two competitive reactions, a second mono-
electron Co(II)/Co(I) transfer and the homolytic disso-
ciation of the CoꢀC bond, as it was previously
described for
a series of organocobaloximes in
dimethylformamide [3]. Even if a numerical value for
this reaction has not been calculated, the increase of the
Ipc(II)/Ipc(I) ratio, at constant CV scan rate [18], indi-
cates that the reaction becomes faster in the order
9:8:7B6B5B10; therefore, the insertion of B(p-
XPh)2 bridges slows the decomposition reaction in
comparison with both the cobaloxime 10 and the com-
plex 5 containing BF2.

F. Asaro et al. / Journal of Organometallic Chemistry 601 (2000) 114–125
125
L.G. Marzilli, Inorg. Chem. 36 (1997) 314; (c) L. Randaccio, N.
Bresciani-Pahor, E. Zangrando, L.G. Marzilli, Chem. Soc. Rev.
18 (1989) 225.
Acknowledgements
We are grateful to CNR (Rome) and to Murst
(Rome) for the financial support.
[8] S. Hirota, S.M. Polson, J.M. Puckett Jr., S.J. Moore, M.B.
Mitchell, L.G. Marzilli, Inorg. Chem. 35 (1996) 5646.
[9] R. Dreos, S. Geremia, G. Nardin, L. Randaccio, G. Tauzher, S.
Vuano, Inorg. Chim. Acta 272 (1998) 74.
[10] G.N. Schrauzer, in: R. Parry (Ed.), Inorganic Syntheses, vol. 11,
1968, p.61 ff.
References
[11] T.P. Povlock, W.T. Lippincott, J. Am. Chem. Soc. 80 (1958)
5409.
[12] (a) G.M. Sheldrick, SHELXS-86, Program for Structure Solution,
Acta Crystallogr. Sect. A 46 (1990) 467; G.M. Sheldrick, Pro-
gram for Structure Refinement, Universita¨t Go¨ttingen, Go¨ttin-
gen, Germany, 1993.
[1] (a) G.N. Schrauzer, Chem. Ber. 95 (1962) 1438; (b) F. Umland,
D. Thierig, Angew. Chem. Int. Ed. Engl. 1 (1962) 333; (c) G.
Schmid, P. Powell, H. No¨th, Chem. Ber. 101 (1968) 1205.
[2] M.S. Ram, C.G. Riordan, G.P.A. Yap, L. Liable-Sands, A.L.
Rheingold, A. Marchaj, J.R. Norton, J. Am. Chem. Soc. 119
(1997) 1648.
[13] J.A. Riddick, in: A. Weissberger, E.S. Proskaue (Eds.), Organic
Solvents. Physical Properties and Methods of Purification,
Wiley–Interscience, New York, 1970, p. XIII.
[14] D. Britz, Electrochim. Acta 25 (1980) 1449.
[15] R. Dreos, G. Nardin, L. Randaccio, G. Tauzher, S. Vuano,
Croat. Chim. Acta 72 (1999) 231.
[16] L.M. Jackman, S. Sternhell, Application of Nuclear Magnetic
Resonance Spectroscopy in Organic Chemistry, Pergamon Press,
Oxford, second Ed., 1969, p. 202.
[17] J. Heyrovsky, J. Kuta, Principles of Polarography, Academic
Press, New York, 1966.
[18] A.J. Bard, L.R. Faulkner, Electrochemical Methods, Wiley, New
York, 1980.
[19] J. Juse´lius, D. Sundholm, Phys. Chem. Chem. Phys. 1 (1999)
3429.
[20] C.A. Hunter, J.K.M. Sanders, J. Am. Chem. Soc. 112 (1990)
5525.
[21] J.N.H. Reek, A.H. Priem, H. Engelkamp, A.E. Rowan,
J.A.A.W. Elemans, R.J.M. Nolte, J. Am. Chem. Soc. 119 (1997)
9956.
[3] (a) C. Tavagnacco, G. Balducci, G. Costa, K. Ta¨schler, W. von
Philipsborn, Helv. Chim. Acta 73 (1990) 1469; (b) G. Costa, A.
Puxeddu, C. Tavagnacco, G. Balducci, R. Kumar, Gazz. Chim.
It. 116 (1986) 735; (c) G. Costa, A. Puxeddu, C. Tavagnacco, J.
Organomet. Chem. 296 (1985) 161 and refs. therein.
[4] (a) F. Asaro, R. Dreos, S. Geremia, G. Nardin, G. Pellizer, L.
Randaccio, G. Tauzher, S. Vuano, J. Organomet. Chem. 548
(1997) 211; (b) R. Dreos, G. Tauzher, S. Vuano, F. Asaro, G.
Pellizer, G. Nardin, L. Randaccio, S. Geremia, J. Organometal.
Chem. 505 (1995) 135.
[5] (a) M. Verhage, D.A. Hoogwater, H. van Bekkum, J. Reedijk,
Recl. Trav. Chim. Pays Bas 101 (1982) 351; (b) D.V. Stynes,
D.B.Leznoff, D.G.A.H de Silva, Inorg. Chem. 32 (1993) 3989;
(c) G. Impey, D.V. Stynes, J. Am. Chem. Soc. 115 (1993) 7868;
(d) D.V. Stynes, Inorg. Chem. 33 (1994) 5022; (e) D.G.A.H. de
Silva, D.B. Leznoff, G. Impey, I. Vernik, Z. Jin, D.V. Stynes,
Inorg. Chem. 34 (1995) 4015; (f) I. Vernik, D.V. Stynes, Inorg.
Chem. 35 (1996) 6210.
[6] S. Shi, L.M. Daniels, J.H. Espenson, Inorg. Chem. 30 (1991)
3407.
[22] A. Bigotto, E. Zangrando, L. Randaccio, J. Chem. Soc. Dalton
Trans. (1976) 96.
[7] (a) N. Bresciani-Pahor, L. Randaccio, E. Zangrando, Inorg.
Chim. Acta 168 (1990) 115; (b) S.M. Polson, R. Cini, C. Pifferi,