An NMR study of the rotational barriers in cobalt-stabilized carbocations: X-ray crystal structures of (η4-C4Ph4)Co-(η5 -C5H4R), where R is CH3C=O, CH=O, CH(tBu)OH

Article abstract of DOI:10.1016/j.jorganchem.2004.02.021

Treatment of the aldehyde (η⁴-C₄Ph₄) Co(η⁵-C₅H₄-CH=O) (4b) with tert-butyllithium or phenyllithium yields the secondary alcohols (η⁴-C₄Ph₄)Co(η⁵- C₅H₄-CH(R)OH), where R=tert-butyl (5) or phenyl (6 . Protonation of 5 and 6 at -80 °C furnishes the deep purple, cobalt-stabilized cations, 7 and 8, respectively, both of which exhibit restricted rotation about the external C₅H₄ -CHR⁺ linkage on the NMR time-scale. These data indicate a minimum value for the barrier to rotation of 15 kcal mol^-1, but it is certainly much higher, indicating a considerable degree of C-C double bond character. X-ray crystal structures of 4b, 5 and also of the ketone (η⁴-C₄Ph₄ Co(η⁵ -C₅H₄-C(=O)CH₃ (4a) are reported. The secondary alcohol 5 exhibits disorder in the solid state because of the presence of diastereomers as a consequence of the stereogenic center at the α-carbon and the clockwise or anticlockwise propeller orientations of the tetraphenylcyclobutadiene ligand.

Full text of DOI:10.1016/j.jorganchem.2004.02.021

Journal of Organometallic Chemistry 689 (2004) 1657–1664
www.elsevier.com/locate/jorganchem
An NMR study of the rotational barriers in cobalt-stabilized
carbocations: X-ray crystal structures of (g4-C4Ph4)Co-(g5-C5H4R),
where R is CH3C@O, CH@O, CH(tBu)OH
a
a
a
a
Yannick Ortin , Kirsten Ahrenstorf , Paul OÕDonohue , Daniel Foede ,
a
Helge Muller-Bunz , Patrick McArdle , Anthony R. Manning
b
a,
*
€
,
a,
*
Michael J. McGlinchey
a
Department of Chemistry, University College Dublin, Belfield, Dublin 4, Ireland
b
Department of Chemistry, NUI, Galway, Ireland
Received 27 January 2004; accepted 20 February 2004
Abstract
Treatment of the aldehyde (g4-C4Ph4)Co(g5-C5H4-CH@O) (4b) with tert-butyllithium or phenyllithium yields the secondary
alcohols (g4-C4Ph4)Co(g5-C5H4-CH(R)OH), where R ¼ tert-butyl (5) or phenyl (6). Protonation of 5 and 6 at )80 °C furnishes the
deep purple, cobalt-stabilized cations, 7 and 8, respectively, both of which exhibit restricted rotation about the external C5H4–
CHRþ linkage on the NMR time-scale. These data indicate a minimum value for the barrier to rotation of 15 kcal molꢀ1, but it is
certainly much higher, indicating a considerable degree of C–C double bond character. X-ray crystal structures of 4b, 5 and also of
the ketone (g4-C4Ph4)Co(g5-C5H4-C(@O)CH3 (4a) are reported. The secondary alcohol 5 exhibits disorder in the solid state be-
cause of the presence of diastereomers as a consequence of the stereogenic center at the a-carbon and the clockwise or anticlockwise
propeller orientations of the tetraphenylcyclobutadiene ligand.
Ó 2004 Elsevier B.V. All rights reserved.
Keywords: Cyclobutadiene–cobalt complexes; Metal-stabilized carbocations; NMR; X-ray crystallography
1. Introduction
site and the relatively electron-rich metal has been
gauged from NMR-derived barriers to migration or
rotation.
The stabilization of carbocationic sites by neigh-
boring transition metal centers continues to attract
widespread attention. Early reports on ferrocenylme-
thylium [1] or (chromium tricarbonyl)benzyl cations
[2], and other related systems [3–6] were subsequently
augmented by extensive studies on cluster-stabilized
cations whose syntheses, structures and molecular
dynamics have been comprehensively reviewed [7–9].
In favorable cases, the structures of M-CRþ2 cations,
where M is ferrocenyl or ruthenocenyl [10], or where
M represents a tetrahedral metal cluster, have been
established X-ray crystallographically [7–9,11], and the
strength of the interaction between the carbocationic
Typically, in [Cp2Mo2(CO)4(RCBC-CR1R2)]þ tetra-
hedral dimetallic clusters (1), the barrier toward migra-
tion of the (C-CR1R2)þ moiety from one molybdenum
vertex to the other ranges from 10 to 18 kcal molꢀ1
depending on the identity of the substituents R1 and R2.
When R1 ¼ R2 ¼ H, the resulting primary cation must
overcome a barrier of 17 kcal molꢀ1 during the fluxional
process that interconverts 1a and 1b [12]. In contrast, for
tertiary cations where R1 and R2 are both alkyl or aryl,
the barrier falls to 10 kcal molꢀ1 [13]. These NMR-de-
rived activation energies correlate very well with the
Mꢁ ꢁ ꢁCþ distances observed crystallographically which
þ
ꢀ
ꢀ
range from 2.44 A for Mꢁ ꢁ ꢁCH2 systems to 2.75 A for
Mꢁ ꢁ ꢁCRþ2 clusters [11]. These data indicate that rela-
tively stable tertiary carbocations have little need for
* Corresponding authors. Tel.: +353-17162880; fax: +353-17161178.
E-mail address: michael.mcglinchey@ucd.ie (M.J. McGlinchey).
0022-328X/$ - see front matter Ó 2004 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2004.02.021

1658
Y. Ortin et al. / Journal of Organometallic Chemistry 689 (2004) 1657–1664
R2
R2
R1
R1
C
C
Cp(CO)2Mo
Mo(CO)2Cp
Cp(OC)2Mo
Mo(OC)2Cp
R
R
1a
1b
Scheme 1.
+
anchimeric assistance from
a
neighboring metal,
R
R
.
.
whereas the positive charge formally associated with a
primary cation is markedly delocalized via direct over-
lap of the vacant p-orbital on carbon with a filled d-
orbital on the metal (see Scheme 1).
.
.
.
.
.
.
R
R
.
.
.
.
Fe
2a
Cr
CO
CO
OC
2b
Similarly, the barrier to tripodal rotation in systems
of the type [(benzyl)Cr(CO)3]þ has been found to be
approximately 11–12 kcal molꢀ1, and once again is a
reflection of the strength of the interaction between
the metal and the carbocationic site [2b,14]. Another
approach is to probe the degree of double bond
character developed between the complexed ring and
the cationic carbon since this will result in an en-
hanced barrier to rotation about the C@CR1R2 link-
age. This approach has recently been used to
demonstrate the ability of a dicobalt–alkyne cluster, or
of a ferrocenyl substituent, to stabilize antiaromatic
fluorenyl, indenyl or cyclopentadienyl cations [15,16].
The particular focus of the current investigation is to
explore the cation-stabilizing ability of systems of the
type [(g4-C4Ph4)Co(g5-C5H4-CHR)]þ and to compare
the results with those for other sandwich compounds.
After much early controversy, there is compelling
experimental and theoretical evidence that ferrocenyl
cations of the type [(C5H5)Fe(C5H4-CHR)]þ are sta-
bilized by direct interaction of the iron atom with the
carbocationic center. Thus, 57Fe NMR data on the
cations and their precursor alcohols exhibit marked
differences in both 57Fe chemical shifts and 13C – 57Fe
coupling constants [17]. Even more convincingly,
X-ray crystallographic results reveal that the exocyclic
carbon can bend down by as much as 25° toward the
metal such that the Feꢁ ꢁ ꢁCþ distance approaches 2.6
+
+
R
R
.
.
.
.
R
.
.
.
.
R
R
.
.
.
.
Co
Mn
2c
R
R
CO
CO
OC
R
2d
It is particularly relevant to cite a study by Gleiter
et al. [19] aimed at probing the similarities and differ-
ences between the diarylferrocenylium ion (C5H5)Fe-
(C5H4-CAr2)]þ and its isolobal cobalt analogue [(C4R4)
Co(C5H4-CAr2)]þ. It was concluded that there is a
greater interaction of the 6,6-diphenylfulvene ligand
with (C5H5)Feþ than with the corresponding
(C4H4)Coþ species, attributable to the enhanced ability
of the latter moiety to distribute positive charge over the
four-membered ring.
We here describe our efforts to provide an experi-
mental value for the barrier to rotation about the
exocyclic bond in [(C4Ph4)Co(C5H4-CHR)]þ, where R
is phenyl or tert-butyl. Such data are, to the best of
our knowledge, not available for cobalt cationic sys-
tems although corresponding values in ferrocenylium
cations may be extracted from early work. Thus, it is
reported that in Fc2CHþ (3), the 13C NMR reso-
nances of the C2;5 environments in the C5H4 rings are
clearly split at )70 °C, but have coalesced at +30 °C
[17b]. These data were acquired on a 90 MHz spec-
trometer (13C at 22.6 MHz), and the Gutowsky–Holm
equation yields an activation energy of approximately
14 kcal molꢀ1. Corresponding data on the Fc–CHMeþ
and Fc–CHPhþ cations reveal that the C2;5 and C3;4
environments do not exhibit coalescence behavior,
even at +15 °C, indicating a minimum barrier of 15
kcal molꢀ1, and probably substantially higher. An
ꢀ
A [10]. An extreme view would regard such complexes
as being comprised of a neutral fulvene ligand coor-
dinated in an g6-fashion to a (C5H5)Feþ moiety, as in
2a. More recently, comparisons have been drawn be-
tween 2a and the analogous chromium, 2b, manga-
nese, 2c, and cobalt, 2d, systems in terms of structural
characteristics, NMR parameters and high level com-
putational data [18].

Y. Ortin et al. / Journal of Organometallic Chemistry 689 (2004) 1657–1664
1659
entirely different approach was adopted by Turbitt
and Watts [20] who separated the enantiomers of Fc–
CH(Me)OH, generated the chiral Fc–CHMeþ cations,
and measured the rate of racemization at several
temperatures to obtain a rotational barrier of 20
by treatment of (C5H4R)Co(CO)2 with diphenylacet-
ylene. However, much better yields can be realized by
reaction of the sodium salt of the appropriately
substituted cyclopentadienide anion, C5H4Rꢀ, with
chloro-tris(triphenylphosphine)cobalt(I), and then in-
corporating the four-membered ring by the coupling
of two alkyne ligands [27] to yield the required com-
plexes, as shown in Scheme 2.
kcal molꢀ1
.
In the diferrocenyl case, 3, the relatively low barrier can
be readily accounted for by the fact that the cation is not
strongly bonded to one iron center but instead is partially
stabilized by each of two metal atoms. Such behavior is
reminiscent of the X-ray crystal structures of {bis[(cy-
clobutadiene)Fe(CO)3]CH}þ [3], and {[Co2(CO)6(tert-
Bu-CBC)]2tert-Bu-C}þ [21], in which the cationic carbon
is displaced only minimally toward both metals and is not
firmly bonded to either.
In order to probe the rotational barriers in systems of
the type [(C4Ph4)Co(C5H4-CHR)]þ one must first de-
velop synthetic routes to their precursor alcohols. Pre-
vious reports have focussed on the readily available
ester, (C4Ph4)Co(C5H4-CO2Me) [22], which, when
treated with Grignard or organo-lithium reagents, yields
the symmetrical alcohols (C4Ph4)Co(C5H4-CR2OH),
where R ¼ H, Me or Ph [23]. Careful oxidation of the
primary alcohol furnishes the aldehyde (C4Ph4)Co
(C5H4-CH@O) [24]. The carbinol (C4Ph4)Co(C5H4-
CPh2OH) had been previously prepared by the reaction
of (C4Ph4)Co(C5H4Li) with benzophenone [4].
Unlike ferrocene, which readily undergoes electro-
philic attack (Friedel–Crafts, Vilsmeier–Haack, etc.)
on the five-membered ring, the cobalt systems are
preferentially substituted, where possible, on the four-
membered ring and functionalization of the five-
membered ring proceeds in very poor yields [25].
Consequently, analogously to the route originally de-
scribed by Rausch and co-workers [26], one must
functionalize the cyclopentadienyl ring prior to coor-
dination by cobalt. The early syntheses of cobalt
sandwiches of the type (C5H4R)Co(C4Ph4) proceeded
2. Results and discussion
Initial synthetic studies focussed on attempts to gen-
erate unsymmetrical tertiary alcohols by treatment of
(C4Ph4)Co(C5H4-COCH3) (4a), with phenyllithium or a
phenyl Grignard reagent. Somewhat surprisingly, this
simple ketone has not, to our knowledge, been charac-
terized X-ray crystallographically, and the structure
appears as Fig. 1. Crystallographic data are collected in
Table 1.
The cobalt to ring-carbon distances average 2.070(2)
ꢀ
ꢀ
A to the cyclopentadienyl ring and 1.982(2) A to the
cyclobutadiene ring; the net result is to place the cobalt
almost equidistant from the two ring centers (Co-to-4-
ꢀ
membered ring 1.682 A; Co-to-5-membered ring 1.692
ꢀ
A). The phenyls adopt a propeller conformation with
dihedral angles ranging from 21° to 42°, and are bent in
an exo fashion out of the plane of the cyclobutadiene
ring by ca. 7.5°. These metric parameters are typical for
sandwich complexes of this type [28].
However, nucleophilic attack on the carbonyl moiety
in 4a with either PhLi or PhMgBr was in all cases less
favorable than proton abstraction from the methyl ke-
tone. The reverse process, i.e. addition of a methyl
Grignard, was thwarted by our inability to prepare
(C4Ph4)Co(C5H4-COPh) in synthetically useful yields,
even though (C5H4-COPh)Co(CO)2 has been previously
reported [29]. However, the secondary alcohols
(C4Ph4)Co(C5H4-CHR-OH), where R is tert-butyl (5),
H
H
-
(Ph3P)3CoCl
O
O
THF
C6
+
Co
C7
C8
C10
C11
Na
O
Ph3P
PPh3
C5
C9
Co
toluene,
reflux
Ph
2
Ph
R
C18
C24
C2
C1
C3
C4
H
H
C19
C12
O
tBuLi or PhLi
THF
OH
Co
Co
C13
C25
C31
C30
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
5 : R = tBu
6 : R = Ph
4b
Scheme 2.
Fig. 1. X-ray crystal structure of (C4Ph4)Co(C5H4-CO-Me) (4a).

1660
Y. Ortin et al. / Journal of Organometallic Chemistry 689 (2004) 1657–1664
Table 1
Crystallographic collection and refinement parameters
Identification code
4a
4b
5
Empirical formula
Formula weight
Temperature (K)
C35H27OCo
522.50
293(2)
C35H27Cl2OCo
593.40
298(2)
C38H35OCo
566.59
293(2)
ꢀ
Wavelength (A)
0.71073
Orthorhombic
P212121 (#19)
0.71069
Monoclinic
P21/c (#14)
0.71073
Monoclinic
P21/c (#14)
Crystal system
Space group
Unit cell dimensions
ꢀ
a (A)
9.0008 (13)
9.9706 (14)
11.3769 (14)
15.022 (3)
11.6707 (8)
9.6187 (7)
ꢀ
b (A)
ꢀ
c (A)
29.868 (4)
90
16.908 (5)
90
26.2564 (18)
90
a (°)
b (°)
c (°)
90
90
81.210 (16)
90
2855.7(10)
92.4570 (10)
90
2944.8(4)
3
ꢀ
Volume (A )
2680.5(7)
4
1.295
Z
4
1.380
4
1.278
Dcalc (Mg/m3)
Absorption coefficient (mmꢀ1
F ð000Þ
)
0.666
1088
0.815
1224
0.611
1192
Crystal size (mm3)
2.00 ꢃ 2.00 ꢃ 1.00
2.15–28.24
0.48 ꢃ 0.43 ꢃ 0.39
1.81–20.89
0.70 ꢃ 0.46 ꢃ 0.30
2.26–27.50
h Range for data collection (°)
Index ranges
ꢀ11 6 h 6 11, ꢀ13 6 k 6 13,
ꢀ39 6 l 6 39
44 796
ꢀ11 6 h 6 11, ꢀ14 6 k 6 14,
ꢀ16 6 l 6 16
ꢀ15 6 h 6 15, ꢀ12 6 k 6 12,
ꢀ34 6 l 6 34
49 162
Reflections collected
11 788
Independent reflections
Completeness for h range
Absorption correction
Max. and min. transmission
Refinement method
6404 [Rint ¼ 0:0320]
98.1%
Numerical
2944 [Rint ¼ 0:0535]
97.3%
None
6772 [Rint ¼ 0:0221]
99.9%
Numerical
0.5557 and 0.3494
Full-matrix least-squares on F2
6404/0/443
Not given
0.8378 and 0.6742
Full-matrix least-squares on F2
6772/0/505
Full-matrix least-squares on F2
2944/0/352
1.006
Data/restraints/parameters
Goodness-of-fit on F2
Final R indices [I > 2rðIÞ]
R indices (all data)
1.088
1.053
R1 ¼ 0:0294, wR2 ¼ 0:0704
R1 ¼ 0:0313, wR2 ¼ 0:0713
0.34 and )0.26
0.010(9)
R1 ¼ 0:0565, wR2 ¼ 0:1496
R1 ¼ 0:0649, wR2 ¼ 0:1627
0.74 and )0.42
–
R1 ¼ 0:0407, wR2 ¼ 0:1076
R1 ¼ 0:0479, wR2 ¼ 0:1119
0.89 and )0.22
–
ꢀ3
ꢀ
Largest diff. peak and hole (e/A
Absolute structure parameter
)
or phenyl (6), were readily preparable from the aldehyde
4b, whose structure appears as Fig. 2. As with the me-
thyl ketone 4a, the metric parameters for the aldehyde
4b are normal. The cobalt is sited almost equidistant
from the two rings, the average twist angle of the phenyl
propeller blades is 40° ꢂ 14°, and the aldehyde func-
tionality is coplanar with the five-membered ring (see
Fig. 2).
The secondary alcohols 5 and 6 were characterized
spectroscopically and by microanalysis, and the for-
mer also yielded crystals suitable for an X-ray dif-
fraction study. The molecular structure of 5 is shown
as Fig. 3 and reveals several interesting features. As
with the previously mentioned ketone, 4a, and alde-
hyde, 4b, the tetraphenylcyclobutadiene ligand in 5
adopts a propeller conformation such that each phenyl
makes a dihedral angle of approximately 39° ꢂ 17°
relative to the four-membered ring, thus rendering the
molecule chiral. However, the compound already
possesses a stereogenic center since the a-carbon bears
four different substituents. The net result is, of course,
to generate diastereomers in which the clockwise or
counterclockwise pitch of the four phenyls could
correlate with either the R or S configuration at Ca.
Each unit cell possesses a crystallographic inversion
C2
C3
C6
C4
C1
C5
O1
Co1
C10
C30
C7
C31
C29
C11
C13
C12
C17
C24
C9
C18
C8
C19
C23
Fig. 2. X-ray crystal structure of (C4Ph4)Co(C5H4-CHO) (4b).

Y. Ortin et al. / Journal of Organometallic Chemistry 689 (2004) 1657–1664
1661
ring indicating that rotation of the C@CHR linkage was
slow on the NMR time-scale. Upon protonation of the
tert-butyl system, 5, the a-carbon is deshielded by 107
ppm (7–5); the corresponding value for the phenyl
complex, 6, is 82 ppm (8–6). The 13C assignments were
C12
C11
C14
C13
1
established via HSQC and HMBC H–13C two-dimen-
C9
C7
sional correlations.
C5
It is now well established that in M-CRþ2 systems
where the R groups can help to stabilize the electronic
deficiency, the 13C NMR shift of the cationic carbon
is very markedly deshielded. Typically, when going
from (C6H5–CPh2OH)Cr(CO)3 to [(C6H5–CPh2)
Cr(CO)3]þ, the a-carbon resonances are found at 79.7
d and 185.8, respectively, a Dd value of 106.1 ppm
[31]. On the other hand, when the majority of the
positive charge are ever more delocalized onto the
organometallic moiety, the 13C resonance of the cat-
ionic carbon is gradually less deshielded. Thus, for the
C10
C8
O1
C6
Co1
C16
C15
C21
C1
C2
C3
C4
C22
C27
C33
C34
C28
alcohol–cation
pairs
(C6H5–CHPhOH)Cr(CO)3/
[(C6H5–CHPhCr(CO)3]þ and (C6H5–CH2OH)Cr(CO)3/
[(C6H5–CH2)Cr(CO)3]þ, the Dd values are 65 ppm
[2b,2c] and ca. 27 ppm [14b], respectively. Similarly,
the Dd values for the alcohol–cation pairs FcCPh2OH/
FcCPhþ2 and FcCHPhOH/FcCHPhþ are 71 and 49
ppm, respectively [19,32].
Fig. 3. X-ray crystal structure of (C4Ph4)Co(C5H4-CHOH-tBu) (5).
center in that the C4Ph4 propellers are enantiomeric.
However, there is an 85:15 disorder of the hydroxyl
and hydrogen substituents such that the R configura-
tion at Ca primarily corresponds to the D propeller
(while, of course, the S configuration favors the K
propeller). Evidently, the crystal packing is dominated
by the chirality of the propellers, and the substituents
buried inside the molecule play only a secondary role.
A similar phenomenon has been previously observed
in (C5Ph5)Fe(CO)(CH@O)PR3 where again an intrin-
sically chiral molecule gives rise to diastereomers when
the propeller orientations are fixed [30].
Having prepared the alcohols (C4Ph4)Co(C5H4-CHR-
OH) (5) and (6), each was treated with HBF4 ꢁ Et2O at dry
ice temperature and the yellow-orange precursors in-
stantly turned deep purple. The 13C NMR spectra of both
7 and 8 exhibited five resonances for the cyclopentadienyl
The observed deshielding of 82 ppm upon proton-
ating (C4Ph4)Co(C5H4-CHPh-OH) (6), to yield the
corresponding cation 8, contrasts with the surprisingly
large Dd value of 107 ppm found for the tert-butyl
analogue 7, which implies a greater interaction of the
metal with the a-PhCHþ unit in 8 than with the t-
BuCHþ moiety in 7. One might have anticipated that
the phenyl groupÕs ability to help delocalize the charge
deficiency might have been more efficient than the
hyperconjugative stabilization provided by the tert-
butyl substituent; evidently, more examples are needed
to clarify this point. 13C NMR data for a series of
related cobalt complexes are collected in Table 2.
When the temperature was gradually raised to )10 °C,
the cyclopentadienyl 13C NMR resonances showed no
evidence of coalescence, but signs of decomposition be-
Table 2
13C NMR data on selected cobalt sandwich complexes
tBu
tBu
Ph
Ph
Ph
H
Ph
Ph
2
1
α
3
H
OH
H
Ph
H
OH
H
Co
Co
Co
Co
Co
Co
13C
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
Ph
4
Ph
a
1
2
3
4
74.6
102.0
70.8
103.7
48.6
106.2
83.8
181.8
111.2
153.0
108.2
184.0
107.7
105.1
90.8
84.0, 83.0
82.0, 81.0
76.4
84.0, 82.9
81.0, 80.2
75.1
114.0, 111.3
92.3, 86.1
86.0
108.9, 107.2
91.4, 84
90.5
83.0
74.8
88.1

1662
Y. Ortin et al. / Journal of Organometallic Chemistry 689 (2004) 1657–1664
came evident for the tert-butyl derivative, 7; the phenyl
cation, 8, survived until +15 °C but still did not exhibit
peak coalescence. These data clearly indicate a ‘‘fulvene-
type’’ structure with a relatively high activation energy
toward rotation of the exocyclic double bond. Although
one cannot directly calculate a barrier, the data yield a
minimum value of ca. 15 kcal molꢀ1 indicating that the
(C4Ph4)Co moiety provides substantial stabilization of a
cationic charge at the a-carbon position. Current studies
are aimed at generating a cation adjacent to both cobalt-
and iron-containing fragments such that a direct compe-
tition can be allowed to develop.
26%). 1H NMR (300 MHz, CDCl3): d 7.51–7.13 (m, Ph),
5.20 and 4.77 (m, Cp), 1.65 (s, CH3). 13C NMR (75.4
MHz, CDCl3): d 197.27 (CO), 135.0 (Cipso, Ph), 128.7
and 128.2 (Cortho and Cmeta, Ph), 126.9 (Cpara, Ph), 93.8,
87.6, 83.2 (Cp), 76.6 (C4Ph4), 27.1 (CH3). IR (THF,
m
CO): 1672 cmꢀ1
.
3.3. (g5-Formylcyclopentadienyl)(g4-tetraphenylcyclo-
butadiene)cobalt (4b)
Using the procedure described in Section 3.2, the al-
dehyde 4b was prepared from sodium cyclopentadie-
nide, ethyl formate, ClCo(PPh3)3 and diphenylacetylene
to yield 4b as a red-brown solid (0.25 g, 0.5 mmol, 9%).
1H NMR (300 MHz, CDCl3): d 9.31 (s, CHO), 7.44–
7.21 (m, Ph), 5.23 and 4.90 (m, Cp). 13C NMR (75.4
MHz, CDCl3): d 191.0 (CO), 135.1 (Cipso, Ph), 129.1 and
128.5 (Cortho and Cmeta, Ph), 127.4 (Cpara, Ph), 108.9,
89.0, 83.4 (Cp), 76.9 (C4Ph4). MS (ES+) m/z 531
[M + Na]þ, 509 [M + H]þ, 415 [M-(HCOC5H4)]þ. IR
3. Experimental
3.1. General methods
All reactions were carried out under an atmosphere
of dry nitrogen. Alumina 90 standardized (Merck) was
used for flash chromatography. NMR spectra were re-
corded on Varian Inova 300 or 500 MHz spectrometers.
Electrospray mass spectrometry was performed on a
Micromass Quattro micro instrument. Infrared spectra
were recorded on a Perkin–Elmer paragon 1000 FT-IR
instrument and were calibrated with polystyrene. Melt-
ing points were determined on an Electrothermal ENG
instrument and are uncorrected. Elemental analyses
were carried out by the Microanalytical Laboratory at
University College Dublin. Solvents were dried by
standard techniques [33].
(THF, mCO): 1684 cmꢀ1
.
3.4. [a-(g4-Tetraphenylcyclobutadiene)cobalt(g5-cyclo-
pentadienyl)]benzyl alcohol (6)
A solution of 4b (0.50 g, 1.0 mmol) in dry THF (15 ml)
was cooled to )60 °C. Phenyllithium (0.53 ml, 1.8M, 0.95
mmol) was slowly added, the temperature of the mixture
rose gradually to 20 °C and it was stirred overnight. The
solution was quenched with water (0.1 ml), the mixture
washed with water and dried. The organic layer was
concentrated in vacuo, and chromatographed on an alu-
mina column. Three bands were eluted using toluene/
hexane mixture; the product was collected in the second
band as a yellow-orange fraction, which upon concen-
tration yielded an orange solid (0.21 g, 0.36 mmol, 36%).
1H NMR (500 MHz, CDCl3): d 7.60–7.11 (m, Ph), 5.16 (d,
3JH;OH ¼ 3 Hz, CH), 4.98, 4.63, 4.57 and 4.55 (m, Cp), 1.64
(d, 3JOH;H ¼ 3 Hz, OH). 13C NMR (125.7 MHz, CDCl3): d
143.5 (Cipso, Ph), 136.2 (Cipso, C4Ph4), 128.3, 128.3 (Cortho
and Cmeta, C4Ph4), 128.3, 126.3 (Cortho and Cmeta, Ph),
127.4 (Cpara, Ph), 126.6 (Cpara, C4Ph4), 103.7 (Cipso, Cp),
84.0, 82.9, 81.0 and 80.2 (CH, Cp), 75.1 (C4Ph4), 70.8
(CHOH). MS (ES+) m/z 569 [M ) OH]þ. Calc. for
C40H31OCo: C, 81.90; H, 5.32; Co, 10.20. Found: C,
81.58; H, 5.40; Co, 9.61%.
3.2. (g5-Acetylcyclopentadienyl)(g4-tetraphenylcyclobut-
adiene)cobalt (4a)
In a 3-necked flask equipped with a condenser, so-
dium metal in excess was added to a solution of freshly
distilled cyclopentadiene (0.50 ml, 7.6 mmol) in dry
THF (25 ml) and the mixture stirred until all the reac-
tion had ceased (ca. 30 min). After removal of the excess
sodium metal, methyl acetate (0.61 ml, 7.7 mmol) was
added to the reaction mixture and the whole was stirred
at reflux for 4 h. During this time a dark red color de-
veloped. After cooling the solution to room tempera-
ture, dry toluene (100 ml) was added, followed by
chlorotris(triphenylphosphine)cobalt (I) (5.0 g, 5.7
mmol). The solution was stirred at room temperature
for 30 min after which time diphenylacetylene (2.0 g,
11.2 mmol) was added, and the mixture was heated at
reflux overnight. The mixture was cooled, filtered
through celite, concentrated in vacuo and chromato-
graphed on an alumina column. Four bands were eluted
using hexane/toluene; the product was collected in the
third band as a dark red fraction, which was concen-
trated to yield a red-brown solid (0.76 g, 1.45 mmol,
3.5. [1-(g4-Tetraphenylcyclobutadiene)cobalt(g5-cyclo-
pentadienyl)]-2,2-dimethyl-propanol (5)
In an analogous fashion to the synthesis of the phenyl
alcohol, 6, the tert-butyl alcohol was prepared by using
t-BuLi rather than PhLi to yield 5 as a yellow solid (0.13
1
g, 0.23 mmol, 79%). H NMR (300 MHz, CDCl3): d
7.48, 7.46 (m, Ph), 4.83, 4.57, 4.51 (m, Cp), 3.68 (d,
3
3JH–H ¼ 3 Hz, CH), 1.08 (d, JH–H ¼ 3 Hz, OH), 0.74

Y. Ortin et al. / Journal of Organometallic Chemistry 689 (2004) 1657–1664
1663
(s, CH3). 13C NMR (75.4 MHz, CDCl3): d 136.3 (Cipso
Ph), 128.7 and 128.3 (Cortho and Cmeta, Ph), 126.5 (Cpara
Ph), 102.0 (Cipso, Cp), 84.0, 83.0, 82.0 and 81.0 (CH,
Cp), 76.4 (C4Ph4), 74.6 (CHOH);,35.4 (CMe3), 26.0
(CMe3). Calc. for C38H35OCo: C, 80.55; H, 6.22; Co,
10.56. Found: C, 77.42; H, 6.22; Co, 10.00%.
,
,
Structures were solved by using the direct-methods
procedure in the Bruker SHELXL [37] program library
and refined by full-matrix least-squares methods on F2.
All non-hydrogen atoms were refined using anisotropic
thermal parameters. Disordered hydrogen atoms were
added as fixed contributors at calculated positions, with
isotropic thermal parameters based on the carbon/oxy-
gen atom to which they are bonded. All other hydrogen
atoms were located in the difference Fourier map and
allowed to refine freely with isotropic temperature fac-
tors. Data for 4b were collected on an Enraf Nonius
CAD4 diffractometer as described in earlier papers [38].
3.6. Protonation of [(C4Ph4)Co(C5H4-CH(t-Bu)-OH)]
A solution of 5 (0.15 g, 0.03 mmol) in CD2Cl2 (0.75
ml) was placed in an NMR tube, under a argon atmo-
sphere, and cooled to )80 °C. A drop of HBF4ꢁOEt2
was then added. The yellow solution immediately turned
purple.
1H NMR (500 MHz, )25 °C, CD2Cl2): d 11.48
(HBF4), 7.62 (CH), 7.52, 7.48 (m, Ph), 6.04, 5.97, 5.84,
5.58 (m, Cp), 4.74 (q, CH2, HBF4.OEt2), 3.78 (q, CH2,
OEt2), 1.59 (t, CH3, HBF4 ꢁ OEt2), 1.28 (t, CH3, OEt2),
0.90 (CH3, t-Bu). 13C NMR (125.7 MHz, )25 °C,
CD2Cl2): 181.8 (CaÞ, 132.6 (Cpara, Ph), 132.4 (Cipso, Ph),
4. Supplementary material
Crystallographic data for the structural analyses have
been deposited at the Cambridge Crystallographic Data
Center, CCDC nos. 229487 (4a), 229489 (4b) and 229488
(5). Copies of this information may be obtained free of
charge from The Director, CCDC, 12 Union Road,
Cambridge CB2 1EZ (Fax: +44-1223-336033; e-mail:
deposit@ccdc.cam.ac.uk or http://www.ccdc.cam.ac.uk.
131.5, 131.3 (Cortho and Cmeta, Ph), 111.2 (Cipso
,
Cp);,114.0, 111.3, 92.3, 86.1 (CH, Cp), 86.0 (C4Ph4),
70.1 (CH2, OEt2); 45.2 (C Me3); 30.0 (CMe3); 16.0 (CH3,
OEt2).
3.7. Protonation of [(C4Ph4)Co(C5 H4-CHPh-OH)]
Acknowledgements
A solution of 6 (0.30 g, 0.05 mmol) in CD2Cl2 (0.75
ml) was placed in an NMR tube, under a argon atmo-
sphere, and cooled to )80 °C. A drop of HBF4 ꢁ OEt2
was then added. The yellow solution immediately turned
Financial support from University College Dublin is
gratefully acknowledged. K.A. thanks the University of
Hamburg for an exchange scholarship.
1
purple. H NMR (500 MHz, )25 °C, CD2Cl2): d 11.32
(HBF4); 7.83, 7.45, 7.30, 7.29 (m, Ph), 7.60 (CH), 6.02,
5.87, 5.82, 5.76 (m, Cp), 4.72 (m, CH2, HBF4.OEt2); 3.71
(m, CH2, OEt2), 1.58 (m, CH3, HBF4 ꢁ OEt2); 1.26 (m,
CH3, OEt2). 13C NMR (125.7 MHz, )25 °C, CD2Cl2): d
153.0 (Ca), 135.5 (Cpara, Ph), 133.7 (Cipso, Ph), 132.7
(Cortho, Ph), 130.8 (Cmeta, Ph), 130.3 (Cipso, C4Ph4), 129.8
References
[1] (a) E.A. Hill, R. Wiesner, J. Am. Chem. Soc. 91 (1969) 509;
(b) R. Gleiter, R. Seeger, H. Binder, E. Fluck, M. Cais, Angew.
Chem., Int. Ed. Engl. 11 (1972) 1028;
(Cpara, C4Ph4), 129.1 and 128.8 (Cortho and Cmeta
,
(c) J.J. Dannenberg, M.K. Levenberg, J.H. Richards, Tetrahe-
dron 29 (1973) 1575.
C4Ph4), 108.2 (Cipso, Cp), 108.9, 107.2, 91.4, 84.2 (CH,
Cp), 90.5 (C4Ph4), 84.1 (CH2, HBF4 ꢁ OEt2); 67.0 (CH2,
OEt2); 14.1 (CH3, OEt2); 12.4 (CH3, HBF4 ꢁ OEt2).
[2] (a) J.D. Holmes, D.A.K. Jones, R. Pettit, J. Organomet. Chem. 4
(1965) 324;
(b) M. Acampora, A. Ceccon, M. Dal Farra, G. Giacometti, G.
Rigatti, J. Chem. Soc., Perkin Trans. 2 (1977) 483;
(c) A. Ceccon, A. Gobbo, A. Venzo, J. Organomet. Chem. 162
(1978) 311;
3.8. Crystallographic data for 4a, 4b and 5
X-ray crystallographic data for 4a and 5 were col-
lected from a suitable sample mounted with grease on
the end of a thin glass fiber. Data were collected on a D8
Bruker diffractometer equipped with a Bruker SMART
APEX CCD area detector (employing the program
SMART) [34] and an X-ray tube utilizing graphite-
(d) D. Seyferth, J.S. Merola, S. Eschbach, J. Am. Chem. Soc. 100
(1978) 4124.
[3] R.E. Davis, H.D. Simpson, N. Grice, R. Pettit, J. Am. Chem. Soc.
93 (1971) 6688.
[4] D. Seyferth, J.S. Merola, J. Organomet. Chem. 160 (1978) 275.
[5] M.D. Rausch, E.A. Mintz, D.W. Macomber, J. Org. Chem. 45
(1980) 689.
ꢀ
[6] (a) M. Cais, Organomet. Chem. Rev. 1 (1966) 435;
(b) L.W. Haynes, R. Pettit, in: G.A. Olah, P.v.R. Schleyer (Eds.),
Carbonium Ions, vol. 5, Wiley, New York, 1976, pp. 2049–2133.
[7] M.J. McGlinchey, L. Girard, R. Ruffolo, Coord. Chem. Rev. 143
(1995) 331.
monochromated Mo Ka radiation (k ¼ 0:71073 A).
Data processing was carried out by use of the program
SAINT [35], while the program SADABS [36] was utilized
for the scaling of diffraction data and an empirical ab-
sorption correction based on redundant reflections.
[8] H. El Amouri, M. Gruselle, Chem. Rev. 96 (1996) 1077.

1664
Y. Ortin et al. / Journal of Organometallic Chemistry 689 (2004) 1657–1664
[9] A.J. Caffyn, K.M. Nicholas, in: G. Wilkinson, F.G.A. Stone,
E.W. Abel (Eds.), Comprehensive Organometallic Chemistry II,
vol. 12, Pergamon Press, Oxford, UK, 1995, pp. 685–702.
[10] (a) S. Lupan, M. Kapon, M. Cais, F.H. Herbstein, Angew.
Chem., Int. Ed. Engl. 11 (1972) 1025;
[20] T.D. Turbitt, W.E. Watts, J. Chem. Soc., Perkin Trans. II (1974)
177.
[21] G.G. Melikyan, S. Bright, T. Monroe, K.I. Hardcastle, J. Ciurash,
Angew. Chem., Int. Ed. Engl. 37 (1998) 161.
[22] S.T. Mabrouk, M.D. Rausch, J. Organomet. Chem. 523 (1996)
111.
(b) U. Behrens, J. Organomet. Chem. 182 (1979) 89;
(c) S. Barlow, A. Cowley, J.C. Green, T.J. Brunker, T. Hascall,
Organometallics 20 (2001) 5351;
[23] (a) M. Uno, K. Ando, N. Komatsuzaki, T. Tanaka, M. Sawada,
S. Takahashi, J. Chem. Soc., Chem. Commun. (1993) 1549;
(b) M. Uno, K. Ando, N. Komatsuzaki, T. Tsuda, T. Tanaka, M.
Sawada, S. Takahashi, J. Organomet. Chem. 473 (1994) 303.
[24] J. Classen, R. Gleiter, F. Rominger, Eur. J. Inorg. Chem. (2002)
2040.
(d) A.Z. Kreindlin, F.M. Dolgushin, A.I. Yanovsky, Z.A.
Kerzina, P.V. Petrovskii, M.I. Rybinskaya, Organomet. Chem.
616 (2000) 106.
[11] L. Girard, P.E. Lock, H. El Amouri, M.J. McGlinchey, J.
Organomet. Chem. 478 (1994) 189.
[25] M.D. Rausch, R.A. Genetti, J. Org. Chem. 35 (1970) 3888.
[26] W.P. Hart, D.W. Macomber, M.D. Rausch, J. Am. Chem. Soc.
102 (1980) 1196.
[12] A. Meyer, D.J. McCabe, M.D. Curtis, Organometallics 6 (1987)
1491.
[13] (a) V.I. Sokolov, I.V. Barinov, O.A. Reutov, Isv. Akad. Nauk
SSSR, Ser. Khim. (1982) 1922;
[27] A.M. Stevens, C.J. Richards, Organometallics 18 (1999) 1346.
[28] C. Benisch, R. Gleiter, T.H. Staeb, B. Nuber, T. Oeser, H.
Pritzkow, F. Rominger, J. Organomet. Chem. 641 (2002) 102.
[29] S.T. Mabrouk, W.P. Hart, M.D. Rausch, J. Organomet. Chem.
527 (1997) 43.
(b) S. Padmanabhan, K.M. Nicholas, J. Organomet. Chem. 268
(1983) C23;
(c) S.F.T. Froom, M. Green, K.R. Nagle, D.J. Williams, J. Chem.
Soc., Chem. Commun. (1987) 1305;
ꢁ
[30] L. Li, A. Decken, B.G. Sayer, M.J. McGlinchey, P. Bregaint, J.-Y.
Thepot, L. Toupet, J.-R. Hamon, C. Lapinte, Organometallics 13
ꢁ
(d) M.V. Galakhov, V.I. Bakhmutov, I.V. Barinov, O.A. Reutov,
J. Organomet. Chem. 421 (1991) 65;
(1994) 682.
(e) C. Cordier, M. Gruselle, G. Jaouen, V.I. Bakhmutov, M.V.
Galakhov, L.L. Troitskaya, V.I. Sokolov, Organometallics 10
(1991) 2303.
[31] K.L. Malisza, L.C.F. Chao, J.F. Britten, B.G. Sayer, G. Jaouen,
S. Top, A. Decken, M.J. McGlinchey, Organometallics 12 (1993)
2462.
[14] (a) M.J. McGlinchey, Adv. Organomet. Chem. 34 (1992) 185;
(b) P.A. Downton, B.G. Sayer, M.J. McGlinchey, Organomet-
allics 11 (1992) 3281.
[32] G.H. Williams, D.D. Traficante, D. Seyferth, J. Organomet.
Chem. 59 (1973) C27.
[33] D.D. Perrin, W.L.F. Armarego, D.R. Perrin, Purification of
Laboratory Chemicals, Pergamon Press, New York, 1980.
[34] G.M. Sheldrick, SMART, Release 4.05, Siemens Energy and
Automation Inc, Madison, WI, 1996.
[15] J.A. Dunn, W.J. Hunks, R. Ruffolo, S.S. Rigby, M.A. Brook,
M.J. McGlinchey, Organometallics 18 (1999) 3372.
[16] L.E. Harrington, I. Vargas-Baca, N. Reginato, M.J. McGlinchey,
Organometallics 22 (2003) 663.
[35] G.M. Sheldrick, SAINT, Release 4.05, Siemens Energy and
Automation Inc, Madison, WI, 1996.
[17] (a) A.A. Koridze, P.V. Petrovskii, S.P. Gubin, E.I. Fedin, J.
Organomet. Chem. 93 (1975) C26;
[36] G.M. Sheldrick, SADABS (Siemens Area Detector Absorption
Corrections), Siemens Energy and Automation Inc, Madison, WI,
1996.
(b) A.A. Koridze, N.M. Astakhova, P.V. Petrovskii, J. Organo-
met. Chem. 254 (1983) 345.
[18] M.A.O. Volland, S. Kudis, G. Helmchen, I. Hyla-Kryspin, F.
Rominger, R. Gleiter, Organometallics 20 (2001) 227.
[37] G.M. Sheldrick, SHELXTL, Version 5.03, Siemens Crystallo-
graphic Research Systems, Madison, WI, 1994.
€
[19] R. Gleiter, H. Schimanke, S.J. Silverio, M. Buchner, G. Huttner,
Organometallics 15 (1996) 5635.
[38] M. Barrow, N.L. Cromhout, D. Cunningham, A.R. Manning, P.
McArdle, J. Renze, J. Organomet. Chem. 563 (1998) 201.