Confirmation of the pentadienyl/alkyne [5 + 2] cycloaddition reactivity of the Cp*Co(η5-1,2,5-trimethylpentadienyl)+complex

Article abstract of DOI:10.1016/j.jorganchem.2016.10.015

The [5 + 2] cycloaddition reactivity of the Cp*Co(η⁵-1,2,5-trimethylpentadienyl)⁺complex and ethyne has been demonstrated. This reactivity was predicted computationally and runs against previous synthetic observations of decreasing reactivity with increasing substitution. The pentadienyl complex and the resulting Cp*Co(η²,η³-1,2,5-trimethylcycloheptadienyl)⁺species have been characterized spectroscopically and crystallographically. The X-ray structure of the pentadienyl complex has a dihedral angle between the two η⁵-bound planes that falls within the range expected for reactivity (3.1(8)°), providing strong support for our computationally proposed structure/reactivity relationship.

Full text of DOI:10.1016/j.jorganchem.2016.10.015

Journal of Organometallic Chemistry 824 (2016) 166e171
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Confirmation of the pentadienyl/alkyne [5 þ 2] cycloaddition
reactivity of the Cp*Co(h
5-1,2,5-trimethylpentadienyl)þ complex
Chandika D. Ramful, Zachary E. Konway, Stephanie Boudreau, Jetsuda Areephong,
Katherine N. Robertson, Kai E.O. Ylijoki*
Department of Chemistry, Saint Mary's University, Halifax, Nova Scotia, B3H 3C3, Canada
a r t i c l e i n f o
a b s t r a c t
Article history:
The [5 þ 2] cycloaddition reactivity of the Cp*Co(
h
5-1,2,5-trimethylpentadienyl)þ complex and ethyne
has been demonstrated. This reactivity was predicted computationally and runs against previous syn-
thetic observations of decreasing reactivity with increasing substitution. The pentadienyl complex and
Received 8 August 2016
Received in revised form
5 October 2016
Accepted 6 October 2016
Available online 8 October 2016
the resulting Cp*Co(h2 3-1,2,5-trimethylcycloheptadienyl)þ species have been characterized spectro-
,h
scopically and crystallographically. The X-ray structure of the pentadienyl complex has a dihedral angle
between the two h5-bound planes that falls within the range expected for reactivity (3.1(8)ꢀ), providing
strong support for our computationally proposed structure/reactivity relationship.
© 2016 Elsevier B.V. All rights reserved.
Keywords:
Cobalt
Pentadienyl
[5 þ 2] cycloaddition
Seven-membered ring
1. Introduction
h
5-pentadienyl substitution patterns have been prepared [5a] and
their [5 þ 2] cycloaddition reactivity evaluated with both acetylene
and 2-butyne [2]. The cycloaddition reactivity of cobalt-
5-penta-
dienyl complexes is highly dependent on the substitution pattern of
the
5-pentadienyl complex. Experimentally, it was observed that
Cycloaddition chemistry has found widespread application in
organic synthesis, with continuous growth observed since the first
reports of the [4 þ 2] Diels-Alder reaction [1]. The modern devel-
opment of metal-mediated cycloaddition mechanisms has
permitted the creation of new reaction pathways that circumvent
orbital symmetry restrictions limiting the application of classic
pericyclic motifs. One example of such a mechanism is the metal-
mediated [5 þ 2] cycloaddition reaction for synthesis of seven-
membered carbocycles [1e]. We have previously reported the
pentadienyl/alkyne [5 þ 2] cycloaddition reaction of half-sandwich
h
h
only substituents at the C1- and C5-positions are tolerated; all
other substitution patterns tested (1,2-, 1,3-, and 1,4-disubstituted,
and 1,2,4-trisubsituted) all proved unreactive [2]. Additionally, a
substituent in the 1-position is mandatory, with the unsubstituted
parent pentadienyl complex proving unreactive [2]. This structure/
reactivity relationship is believed to arise from a combination of h5
-
pentadienyl complex distortion and transition state strain. A
dihedral angle between the two
Cp*Co(
complexes
conjugated h2
h
5-pentadienyl) cations 1 to yield h2
,
h
3-cycloheptadienyl
h
5-bound planes of less than ca.
2
under kinetic control (Scheme 1) [2,3]. Non-
7.6ꢀ is required for reactivity, with substituents in both the 1- and
5-positions producing the greatest effect by enforcing angles of less
than 4ꢀ [4]. With the common theme in the synthetic studies being
an observed decrease in reactivity with increasing substitution, we
were surprised to find that density functional theory predicted
,
h
3-cycloheptadienyl intermediates wherein R1 ¼ H
isomerize to the thermodynamically favoured conjugated h5
-
cycloheptadienyl complexes 3 at elevated temperature. This isom-
erization is believed to occur via dissociation of the
2-alkene with
concomitant formation of an agostic hydride complex; -hydride
elimination and reinsertion complete isomerization [4]. When R1
and R2 s H, -hydride elimination is blocked, rendering the h2 h3
h
b
Cp*Co(h
5-1,2,5-trimethylpentadienyl)þ (1a) to be highly reactive
towards acetylene [4]. Both the calculated transition state barrier
for rate-limiting alkyne capture (23.9 kcal/mol) and the dihedral
angle between the h5-planes (3.67ꢀ) were in the predicted range for
b
, -
cycloheptadienyl complexes thermally stable [2b]. A wide range of
effective reactivity. Here we report our synthesis of Cp*Co(h
5-1,2,5-
trimethylpentadienyl)þ complex 1a and its reactivity with alkynes,
providing strong support for our mechanistic proposal and struc-
ture/reactivity model.
* Corresponding author.
E-mail address: kai.ylijoki@smu.ca (K.E.O. Ylijoki).
http://dx.doi.org/10.1016/j.jorganchem.2016.10.015
0022-328X/© 2016 Elsevier B.V. All rights reserved.

C.D. Ramful et al. / Journal of Organometallic Chemistry 824 (2016) 166e171
167
Scheme 3. Preparation of nonconjugated pentadienol 5 as a mixture of stereoisomers.
Scheme 1.
h
5-Pentadienyl/alkyne [5 þ 2] cycloaddition.
2. Results and discussion
We have previously reported the HBF4$OEt2-mediated synthesis
of Cp*Co(h 4
5-pentadienyl)þ-type complexes 1 from Cp*Co(C2H4)2
with both nonconjugated (Scheme 2, Route A) and conjugated
pentadienols (5 and 6) [5a,6]. We first explored the preparation of
Cp*Co(h
5-1,2,5-trimethylpentadienyl)þ 1a from the nonconjugated
(2E,5E)-3-methylhepta-2,5-dien-4-ol (5). As a trial synthesis we
prepared 2-bromo-2-butenyl magnesium bromide from a 1:1
mixture of 2-bromo-2-butene stereoisomers and subsequently
quenched with (E)-but-2-enal (Scheme 3). The resulting 3-
methylhepta-2,5-dien-4-ol product was formed as a 4:1 mixture
of stereoisomers favouring the undesired (2Z,5E)-3-methylhepta-
2,5-dien-4-ol. During Grignard reagent formation it is likely that a
radical intermediate isomerized to the less sterically hindered vinyl
magnesium halide product, which was subsequently trapped with
the aldehyde [7]. The enrichment with the undesired isomer sug-
gests that a stereochemically pure starting material (i.e., (E)-2-
bromobut-2-ene) will prove equally problematic, and we there-
fore turned our attention to preparation of the conjugated (3E,5E)-
5-methylhepta-3,5-dien-2-ol (6) (Scheme 4). Sodium hydride-
mediated Horner-Emmons-Wadsworth reaction between triethyl-
phosphonoacetate and tiglic aldehyde in THF first gave (2E,4E)-
ethyl 4-methylhexa-2,4-dienoate (7) [8]. Unexpectedly, the use of
NaH as a suspension in mineral oil is essential for clean reactivity; if
the suspension is first washed with hexane, as is commonly done,
the reaction yields a complex mixture of unidentified products.
Subsequent LiAlH4 reduction yielded alcohol 8, which was then
oxidized with activated MnO2 on carbon to provide aldehyde 9
[9,10]; addition of methyl magnesium bromide produces alcohol 6,
completing the synthetic sequence.
Scheme 4. Preparation of conjugated pentadienol 6.
entailed olefin exchange at elevated temperature, followed by
HBF4$OEt2 protonation at ꢁ78 ꢀC. This procedure was intended to
improve the yield of heavily substituted h
5-pentadienyl substrates,
but little difference was observed [5a]. Therefore, to reduce pro-
cedural complexity, we chose to investigate pentadienyl synthesis
analogous to Route A with conjugated dienols (Scheme 2, Route B).
We were pleased to find that the reaction proceeds smoothly, giv-
ing a 78% yield of pentadienyl complex 1a, which is in line with
yields obtained for related complexes (the corresponding reaction
from the pentadienol mixture 5 gave a poor yield of 20%). Crys-
tallization via slow diffusion of diethyl ether into methylene chlo-
ride resulted in crystals suitable for X-ray diffraction. The solid state
structure agrees well with that proposed computationally, with the
key dihedral angle indicating a higher degree of co-planarity than
predicted, measuring 3.1(8)ꢀ compared to a calculated value of
3.67ꢀ (Fig. 2) [4].
With the requisite pentadienyl complex in hand we turned our
efforts towards its [5 þ 2] cycloaddition reactivity (Scheme 5).
Pentadienyl complex 1a was dissolved in methylene chloride in a
test tube and acetylene gas was bubbled through the system until
saturation. The tube was sealed with a septum and allowed to react
overnight. Removal of solvent and benchtop chromatography pro-
Our original protocol for the preparation of Cp*Co(h5
pentadienyl)þ complexes from conjugated pentadienol substrates
-
vided the h2 3-cycloheptadienyl complex 2a in 87% yield as evi-
,h
denced by NMR spectroscopy and X-ray crystallography.
Cycloheptadienyl complex 2a is stable at moderately elevated
temperature (60e70 ꢀC), similar to the analogous Cp*Co(h2 3-1,5-
,h
cycloheptadienyl)þ complex previously reported [2b]. We also
Scheme 2. Two routes for [Cp*Co(
h
5-1,2,5-trimethylpentadienyl)]BF4 (1a) preparation.
Scheme 5. Preparation of [Cp*Co(h2 3-1,2,5-trimethylcycloheptadienyl)]BF4 (2a).
,h

168
C.D. Ramful et al. / Journal of Organometallic Chemistry 824 (2016) 166e171
investigated [5 þ 2] cycloaddition with 2-butyne. Upon addition of
2-butyne and heating, we were surprised to find that the expected
h2 3-cycloheptadienyl complex was not obtained, but rather con-
,h
version to an unidentified product mixture had occurred. This
product instability is likely due to the high steric bulk of the h2 h3
,
-
pentamethylcycloheptadienyl intermediate favouring ligand
dissociation; similar difficulty occurs with the 2-butyne adduct of
the Cp*Co(h
5-1,5-dimethylpentadienyl)þ complex, with slow
decomposition being observed in solution at room temperature
[2b].
3. Crystallography
Pentadienyl complex 1a was obtained as dark red rectangular
needles through two-chambered liquid diffusion of Et2O into a
CH2Cl2 solution (Fig. 1). Crystallographic details are compiled in
Table 1. The crystal was multiply twinned and the BFꢁ4 ion was
disordered. After the final refinement, residual electron density
peaks were observed near the two carbon termini of the open
pentadienyl ligand. These were each about 1 Å from the carbon
atom and 0.3e0.4 eÅꢁ3 in height. They appeared to be significant
(though weaker than the main H peaks) yet a model could not be
found to properly account for them. Whether the hydrogen atoms
at the carbon termini were geometrically placed or refined, the
same Fourier peaks were still observed.
Fig. 2. X-ray crystal structure of [Cp*Co(h2
(2a) with non-hydrogen atoms represented by thermal ellipsoids drawn at the 20%
probability level. Hydrogen atoms, the disordered [BF4ꢁ
anions, and the minor
component of the disorder in the trimethylcycloheptadienyl ligand have been omitted
for clarity. Final residuals: R1 ¼ 0.0509, wR2 ¼ 0.1559. Selected interatomic distances
(Å): Co-C12, 2.115(10); Co-C13, 2.106(9); Co-C15, 2.199(13); Co-C16, 2.057(16); Co-C17,
2.198(11); C11-C12, 1.487(15); C12-C13, 1.346(16); C13-C14, 1.539(19); C14-C15,
1.518(18); C15-C16, 1.394(17); C16-C17, 1.371(15); C11-C17, 1.481(13); C11-C18,
1.493(14); C14-C20, 1.497(16); C15-C19, 11.530(19). Selected angles (ꢀ):C11-C12-C13,
127.5(12); C12-C13-C14, 120.0(12); C13-C14-C15, 102.1(9); C14-C15-C16, 124.8(13);
C15-C16-C17, 126.0(14); C11-C17-C16, 129.2(12).
,h
3-1,2,5-trimethylcycloheptadienyl)][BF4]
]
The Cp*Co(h2 3-1,2,5-trimethylcycloheptadienyl)þ complex 2a
,h
was obtained as deep red plates through two-chambered liquid
diffusion of Et2O into a CH2Cl2 solution (Fig. 2). The structure was
significantly disordered, with the BFꢁ4 anion and the h2 3-cyclo-
,h
heptadienyl ligand being involved. The boron atoms of both anions
are located on special positions. In one anion, the disorder was best
described as an equal contribution from two orientations, while in
the second anion refinement of the occupancy factor gave a 73/27
ratio of the two possible orientations. The seven-membered ring of
the ligand was also modelled with a two-fold disorder. Refinement
of the occupancy gave a 70/30 ratio of the two configurations. The
ring can be coordinated to the cobalt through its two opposing
faces, such that each orientation is numbered counter clockwise to
the other (Fig. 3) [11].
Table 1
X-ray crystallographic details.
1a
2a
formula
fw
C18H28BCoF4
390.14
C20H30BCoF4
416.18
0.350 ꢂ 0.200 ꢂ 0.190
orthorhombic
Pnn2 (No. 34)
Cryst. dimens. (mm)
Cryst. system
space group
unit cell param.
a (Å)
0.280 ꢂ 0.070 ꢂ 0.040
monoclinic
P21/c (No. 14)
7.6504(13)
15.127(3)
15.958(3)
96.843(2)
1833.7(5)
4
1.413
0.970
21243
4470
14.720(2)
15.047(2)
8.9487(13)
b (Å)
c (Å)
b
(deg)
V (Å3)
1982.1(5)
4
Z
rcalcd (Mg mꢁ3
)
1.395
0.902
23192
4929
3223
1.018
m
(mmꢁ1
)
total data
no. indep. reflns.
no. obsd. reflns.
goodness-of-fit (S)
final R indices
R1
3403
1.033
0.0452
0.0509
wR2
0.1093
0.437, ꢁ0.395
0.1559
0.451, ꢁ0.215
Drmax
,
Drmin (e Åꢁ3
)
4. Conclusion
We have demonstrated the [5 þ 2] cycloaddition reactivity of
Cp*Co(
5-1,2,5-trimethylpentadienyl)þ 1a with acetylene,
h
a
computational prediction that ran against experimental observa-
tionsofdecreasedreactivitywithincreasedpentadienylsubstitution.
While cycloaddition with acetylene proceeded as expected to yield
Fig. 1. X-ray crystal structure of [Cp*Co(h
5-1,2,5-trimethylpentadienyl)][BF4] (1a) with
non-hydrogen atoms represented by thermal ellipsoids drawn at the 20% probability
level. Hydrogen atoms and the disordered [BFꢁ4 ] anion have been omitted for clarity.
Final residuals: R1 ¼ 0.0452, wR2 ¼ 0.1093. Selected interatomic distances (Å): Co-C1,
2.160(3); Co-C2, 2.097(3); Co-C3, 2.047(3); Co-C4, 2.052(3); Co-C5, 2.219(3); C1-C2,
1.397(5); C2-C3, 1.403(5); C3-C4, 1.445(5); C4-C5, 1.381(5); C1-C6, 1.518(4); C2-C7,
1.522(4); C5-C8, 1.506(4). Selected angles (ꢀ): C1-C2-C3, 121.8(3); C2-C3-C4, 125.5(3);
C3-C4-C5, 127.9(3).
the Cp*Co(h2 3-1,2,5-trimethylcycloheptadienyl)þ complex, reac-
,h
tion with 2-butyne led to an unidentified product mixture, possibly
due to decomposition of the sterically hindered putative
Cp*Co(h2 3-1,2,3,4,5-pentamethylcycloheptadienyl)þ intermediate.
,h

C.D. Ramful et al. / Journal of Organometallic Chemistry 824 (2016) 166e171
169
,h
Fig. 3. X-ray crystal structure of [Cp*Co(h2 3-1,2,5-trimethylcycloheptadienyl)][BF4] (2a) drawn to show the contributions of the two different ligand conformations. On the left is
the major conformer (70%) on the right is the minor (30%). Thermal ellipsoids of the non-hydrogen atoms are drawn at the 20% probability level.
5. Experimental
5.61 (dd, J ¼ 15.7, 6.6 Hz, 1H), 5.58 (q, J ¼ 6.6 Hz, 1H), 4.37 (quint,
J ¼ 6.4 Hz, 1H), 1.75 (s, 3H), 1.74 (d, 3H, obscured by other signal),
Reagents and Methods. All manipulations on air and moisture
sensitive compounds were performed under nitrogen atmosphere
using standard Schlenk techniques or in a glovebox. Acetone was
dried over boric oxide, degassed via three freeze-pump-thaw cy-
cles, vacuum-transferred and stored under nitrogen. Diethyl ether
was dried by passing through activated Al2O3 (mBraun SPS). All
other reagents were used without further purification. Flash col-
1.63 (br. s, 1H), 1.31 (d, J ¼ 6.4 Hz, 3H). 13C{1H} NMR (75 MHz,
CDCl3): d 134.6,133.8, 130.1, 127.3, 69.2, 23.5,13.8,12.0. Electrospray
MS m/z calculated for C8H15O (M þ Hþ): 127.1123; found: 127.0375.
[Cp*Co(h
5-1,2,5-trimethylpentadienyl)]þBF4¡ (1a). A Schlenk
tube was charged with a solution of Cp*Co(C2H4)2 (135 mg,
0.54 mmol) in acetone (~5 mL) in a glove box and equipped with a
stir-bar and septum. The solution was removed to a Schlenk line,
placed under a nitrogen atmosphere, and cooled to ꢁ78 ꢀC. Tetra-
umn chromatography was performed with silica gel (40e63 mm). IR
spectra were recorded on a Brüker Platinum ATR with diamond
crystal. 1H and 13C NMR spectra were recorded on a Bruker
Avance300 (1H, 300 MHz; 13C, 75 MHz) spectrometer. 1H NMR
chemical shifts are reported relative to residual protiated solvent.
13C NMR chemical shifts are reported relative to the deuterated
solvent. High-resolution mass spectra, performed by Mr. Xiao Feng
of the DalChem Mass Spectrometry Laboratory at Dalhousie Uni-
versity, were obtained on a Brüker microTOF Focus Mass Spec-
trometer. Elemental analysis was performed by the Centre for
Environmental Analysis and Remediation at Saint Mary's University
using a Perkin Elmer 2400 Series II CHN Analyzer. Activated MnO2
[10], (2E,4E)-4-methylhexa-2,4-dien-1-ol (8) [9], (2E,4E)-4-
methylhexa-2,4-dienal (9) [9], and Cp*Co(C2H4)2 (4) [12] were
prepared according to literature procedures. (2E,4E)-Ethyl 4-
methylhexa-2,4-dienoate (7) was prepared using a modification
to a literature procedure by substituting unwashed 60% NaH sus-
pension in mineral oil for KOtBu and purifying by silica gel chro-
matography using a gradient column of hexane, 1% EtOAC/hexanes,
then 2% EtOAc/hexanes to ensure removal of the mineral oil,
otherwise all procedural details were identical [8].
fluoroboric acid (54% in diethyl ether, 74 mL, 0.54 mmol) was added
by syringe and the reaction was stirred at ꢁ78 ꢀC for 10 min. A
solution of (3E,5E)-5-Methylhepta-3,5-dien-2-ol (6) (68 mg,
0.54 mmol) in acetone (1 mL) was then added and the reaction was
allowed to warm gradually to room temperature overnight (~16 h).
The solvent was removed in vacuo and the crude product was pu-
rified by silica gel chromatography, eluting with a 3% methanol in
dichloromethane mixture. Removal of solvent in vacuo afforded 1a
as a red solid (165 mg, 78%) that was of suitable purity for use in
subsequent reactions. Crystals suitable for diffraction analysis were
grown via two-chambered liquid diffusion (Et2O into CH2Cl2).
Compound 1a decomposes slowly in solution, therefore only
spectroscopic characterization was obtained. IR (filmcast, cmꢁ1):
2967 (w), 2915 (w), 1474 (w), 1455 (w), 1429 (w), 1386 (m), 1092
(m), 1048 (s), 1031 (s). 1H NMR (300 MHz, acetone-d6):
d 6.41 (d,
J ¼ 7.24 Hz, 1H), 5.14 (dd, J ¼ 12.7, 7.0 Hz, 1H), 2.78 (dq, J ¼ 12.7,
6.4 Hz, 1H), 2.20 (s, 3H), 1.87 (s, 15H), 1.52 (d, J ¼ 6.3 Hz, 3H), 1.42 (d,
J ¼ 6.4 Hz, 3H), 1.26 (m, 1H). 1H-1H COSY (300 MHz, acetone-d6):
d
6.41 4
d
5.14;
d
5.14 4
d
2.78;
d
2.78 4
d 1.42; d 1.52 4 d 1.26.
13C{1H} NMR (75 MHz, CDCl3):
d
110.4, 99.1, 96.5, 94.1, 87.2, 73.7,
(3E,5E)-5-methylhepta-3,5-dien-2-ol (6). In a Schlenk flask
under nitrogen, (2E,4E)-4-methylhexa-2,4-dienal (9) (1.05 g,
8.6 mmol) was dissolved in THF (50 mL) and cooled to ꢁ78 ꢀC in a
dry ice/acetone bath. To this solution, 3.0 M methyl magnesium
bromide (3.45 mL, 10.3 mmol) was added via syringe. The solution
was stirred for 30 min at ꢁ78 ꢀC, then allowed to warm to room
temperature. The reaction was quenched with saturated NH4Cl
solution, extracted with diethyl ether, washed with brine, and dried
over MgSO4. The solvent was removed under vacuum and the crude
material purified by silica gel column chromatography with 20%
EtOAC/hexane as eluent to yield 650 mg (54%) of product as a pale
yellow oil. IR (neat, cmꢁ1): 3347, 2970, 2922, 2861,1650, 1448, 1394,
1379, 1367, 1350, 1331, 1300, 1146, 1082, 1056, 1027, 962, 940, 870,
19.1, 17.4, 14.8, 9.2. Electrospray MS m/z calculated for C18H28Co
(Mþ): 303.1523; found: 303.1504.
[Cp*Co(h2
,h
3-1,2,5-trimethylcycloheptadienyl)]þBF4¡
(2a).
mg,
[Cp*Co(
h
5-1,2,5-trimethylpentadienyl)]þBF4ꢁ
(1a)
(165
0.42 mmol) was dissolved in CH2Cl2 (15 mL) in a test tube. Acety-
lene gas was then bubbled through the solution for 15 min in order
to ensure saturation. The tube was sealed with a rubber septum and
allowed to stand at room temperature overnight. The solvent was
then removed in vacuo and the product purified by silica gel
chromatography using 3% methanol in dichloromethane. The red
fraction was collected and dried providing 153 mg (87%) of product
as a thick, red oil. Crystals suitable for diffraction analysis were
grown via two-chambered liquid diffusion (Et2O into CH2Cl2). IR
(filmcast, cm-1): 2964 (w), 2918 (w), 2876 (w), 1456 (w), 1431 (w),
853, 843, 797. 1H NMR (300 MHz, CDCl3):
d
6.23 (d, J ¼ 15.7 Hz, 1H),

170
C.D. Ramful et al. / Journal of Organometallic Chemistry 824 (2016) 166e171
1384 (w), 1281 (w), 1260 (w), 1090 (m), 1045 (s), 1026 (s), 865 (w),
800 (m). 1H NMR (300 MHz, CDCl3):
so was omitted from the final refinement. The structure was refined
as a racemic twin to obtain a value for the Flack parameter with an
accurate estimated standard deviation. The value obtained, 0.04(4),
suggests that the absolute configuration chosen is the correct one
and that this crystal is in fact not a twin. The BeF bond lengths in
the disordered anions were restrained to be equal, as were the F/F
distances. The SAME instruction was used in the SHELXL refine-
ment to restrain the two rings to have similar conformations. In
addition, specific bonds in the two rings were restrained to be
equal. Finally, one carbon atom in each ring (C18) was ill-behaved
and restrained to make its motion more isotropic.
d
3.53 (dd, J ¼ 7.8, 3.3 Hz, 1H),
3.36 (d, J ¼ 7.8 Hz, 1H), 2.45e2.25 (m, 4H), 1.81 (s, 18H, overlapping
Cp* and Me peaks), 1.50 (d, J ¼ 6.6 Hz, 3H), 1.39 (d, J ¼ 6.9 Hz, 3H).
13C{1H} NMR (75 MHz, CDCl3):
d 97.0, 90.4, 67.2, 54.3, 52.5, 50.1,
30.8, 27.3, 19.9, 19.2, 16.8, 9.6. Electrospray MS m/z calculated for
C
C
20H30Co (Mþ): 329.1680; found: 329.1676. Analysis calculated for
20H30CoBF4: C, 57.72; H, 7.27; found: C, 57.41; H, 7.16.
Crystallographic Experimental Details. X-ray structure mea-
surements were made on a Bruker APEXII CCD equipped diffrac-
tometer (30 mA, 50 kV) using monochromated Mo K radiation
¼ 0.71073 Å) at 125 K [13]. The crystal chosen for each analysis
was attached to the tip of a 400 m MicroLoop with paratone-N oil.
a
(l
m
Acknowledgments
For each determination, the initial orientation and unit cell were
indexed using a least-squares analysis of a random set of reflections
Financial support from the Canada Foundation for Innovation
(CFI), the Faculties of Science and Graduate Studies and Research of
Saint Mary's University, ACEnet (ACEnet Summer Research
Fellowship to C.D.R.), and the Strategic Cooperative Education
Incentive (summer co-op funding for Z.E.K.) is gratefully
acknowledged.
collected from three series of 0.5ꢀ
u
-scans, 15 s per frame and 12
frames per series, that were well distributed in reciprocal space. For
data collection, four
-scan frame series were collected with 0.5ꢀ
u
wide scans, 30 s frames and 366 frames per series at varying 4
angles (4 ¼ 0ꢀ, 90ꢀ, 180ꢀ, 270ꢀ). The crystal to detector distance was
set to 6 cm and a complete sphere of data was collected. Cell
refinement and data reduction were performed with the Bruker
SAINT [14] software, which corrects for beam inhomogeneity,
possible crystal decay, Lorentz, and polarisation effects. A multi-
scan absorption correction was applied (SADABS [15]). The struc-
tures were solved using SHELXS-2014 [16] (1a) or SHELXT-2014
[16] (2b) and were refined using a full-matrix least-squares
method on F2 with SHELXL-2014 [16]. The non-hydrogen atoms
were refined anisotropically. Hydrogen atoms bonded to carbon
were included at geometrically idealized positions and were not
refined. The isotropic thermal parameters of the hydrogen atoms
were fixed at 1.2Ueq of the parent carbon atom (1.5Ueq was used for
methyl hydrogens). All diagrams were prepared using the programs
POV-Ray 3.7 [17] via Diamond 3.2 [18] or Mercury CSD 3.7 [19].
Compound 1a: The crystal chosen for data collection was found to
be multiply twinned during the data processing and refinement.
The original unit cell chosen was orthorhombic C-centred, however,
there was no space group that matched the observed systematic
absences. If instead a primitive monoclinic cell was selected (with
the original B-axis halved and the B- and C- axes transformed) the
structure could be solved in P21 but with very poor results. Platon
[20] suggested that there was still unaccounted for symmetry in the
structure and halving the C-axis again gave the still smaller
monoclinic cell finally used for the refinement. After taking care of
the pseudo-merohedral twinning, the refinement in P21/c pro-
ceeded smoothly but gave statistical parameters that were higher
than expected. The routine TwinRotMat in the program Platon [20]
was used to find the correct twin law for this non-merohedral
twinning and to write the HKLF5 file (data merged) used in the
refinement. BASF refined to 0.310(1). The fluorine atoms of the BF4ꢁ
anion were disordered. Each fluorine atom was split over two po-
sitions (A and B), with the BeF bond distances in the anions
restrained to be equal and the thermal parameters of each fluorine
atom weakly restrained to be similar. One of the fluorine atoms also
had restraints placed on it to make its thermal parameters more
isotropic. The occupancies of the A and B positions refined to
88.5(7)% and 11.5% respectively. Even after modelling the disorder,
there remained a B level and a C level checkcif alert concerning
large Hirshfeld differences along some of the BeF bonds. Nothing
more was done about these; they arise only because of the large
anisotropic motion of the fluorine ligands, even in the disordered
model. Compound 2a: The RIGU command was included in the
SHELXL refinement to add enhanced rigid bond restraints to the
atomic displacement parameters for 1,2- and 1,3- heavy atom pairs.
One reflection (1 1 0) was partially obscured by the beam stop and
Supplementary material
The cif files CCDC-1496204 (1a) and CCDC-1496205 (2a) have
been deposited as Supplementary Material. These data can be ob-
tained free of charge from the Cambridge Crystallographic Data
Centre via www.ccdc.cam.ac.uk/data_request/cif.
References
[1] (a) A.R. Katritzky, N. Dennis, Chem. Rev. 89 (1989) 827e861;
(b) K.V. Gothelf, K.A. Jorgensen, Chem. Rev. 98 (1989) 863e910;
(c) G.W. Goodall, W. Hayes, Chem. Soc. Rev. 35 (2006) 280e312;
(d) M. Harmata, Chem. Commun. 46 (2010) 8886e8903;
(e) M. Harmata, Chem. Commun. 46 (2010) 8904e8922;
(f) A.G. Lohse, R.P. Hsung, Chem. Eur. J. 17 (2011) 3812e3822;
(g) A. Moyano, R. Rios, Chem. Rev. 111 (2011) 4703e4832;
(h) K.E.O. Ylijoki, J.M. Stryker, Chem. Rev. 113 (2013) 2244e2266.
[2] (a) R.D. Witherell, K.E.O. Ylijoki, J.M. Stryker, J. Am. Chem. Soc. 130 (2008)
2176e2177;
€
(b) K.E.O. Ylijoki, A.D. Kirk, S. Bocklein, R.D. Witherell, J.M. Stryker, Organo-
metallics 34 (2015) 3335e3357.
[3] Related
h
5-pentadienyl/alkyne reactions: (a) A.M. Wilson, T.E. Waldman,
A.L. Rheingold, R.D. Ernst, J. Am. Chem. Soc. 114 (1992) 6252e6254;
(b) C.G. Kreiter, E.C. Koch, W. Frank, G. Reiss, Inorg. Chim. Acta 220 (1994)
95e100;
(c) C. Wang, J.B. Sheridan, H.J. Chung, M.L. Cote, R.A. Lalancette, A.L. Rheingold,
J. Am. Chem. Soc. 116 (1994) 8966e8972;
(d) C.G. Kreiter, C. Fiedler, W. Frank, G.J. Reiss, Chem. Ber. 128 (1995)
515e518;
(e) C.G. Kreiter, C. Fiedler, W. Frank, G.J. Reiss, J. Organomet. Chem. 490 (1995)
125e131;
(f) C.G. Kreiter, E.C. Koch, W. Frank, G.J. Reiss, Z. Naturforsch 51B (1996)
1473e1485;
(g) W. Chen, H.-J. Chung, C. Wang, J.B. Sheridan, M.L. Cote, R.A. Lalancette,
Organometallics 15 (1996) 3337e3344;
(h) H.-J. Chung, J.B. Sheridan, M.L. Cote, R.A. Lalancette, Organometallics 15
(1996) 4575e4585;
(i) R. Tomaszewski, I. Hyla-Kryspin, C.L. Mayne, A.M. Arif, R. Gleiter, R.D. Ernst,
J. Am. Chem. Soc. 120 (1998) 2959e2960;
(j) R. Basta, B.G. Harvey, A.M. Arif, R.D. Ernst, Inorg. Chim. Acta 357 (2004)
3883e3888;
ꢀ
ꢀ
ꢀ
(k) B.G. Harvey, C.L. Mayne, A.M. Arif, R. Tomaszewski, R.D. Ernst, J. Am. Chem.
Soc. 127 (2005) 16426e16435;
(l) B.G. Harvey, A.M. Arif, R.D. Ernst, J. Organomet. Chem. 691 (2006)
5211e5217.
[4] K.E.O. Ylijoki, P.H.M. Budzelaar, J.M. Stryker, Chem. Eur. J. 18 (2012)
9894e9900.
[5] (a) K.E.O. Ylijoki, R.D. Witherell, A.D. Kirk, S. Bocklein, V.A. Lofstrand,
€
R. McDonald, M.J. Ferguson, J.M. Stryker, Organometallics 28 (2009)
6807e6822;
Other Co(III) h
5-pentadienyl complex preparations: (b) P. Powell, L.J. Russel, J.
Chem. Res. 283 (1978) 3652e3672;
(c) R.D. Ernst, H. Ma, G. Sergeson, T. Zahn, M.L. Ziegler, Organometallics 6
(1987) 848e853;

C.D. Ramful et al. / Journal of Organometallic Chemistry 824 (2016) 166e171
171
(d) R.B. Cracknell, J.C. Nicholls, J.L. Spencer, Organometallics 15 (1996)
446e448;
[11] For a similar disordered system see the X-ray structure for compound 34 in
Ref. [2b].
(e) M.V. Butovskii, U. Englert, G.E. Herberich, K. Kirchner, U. Koelle, Organo-
metallics 22 (2003) 1989e1991;
[12] (a) C.D. Ramful, K.N. Robertson, K.E.O. Ylijoki, Acta Cryst. E72 (2016)
1301e1304;
Other low valent cobalt
h
5-pentadienyl complex preparations: (f) J.R. Bleeke,
Also see: (b) J.C. Nicholls, J.L. Spencer, Inorg. Synth. 28 (1990) 278e280;
(c) R.G. Beevor, S.A. Frith, J.L. Spencer, J. Organomet, Chem. 221 (1981)
C25eC27.
W.-J. Peng, Organometallics 3 (1984) 1422e1426;
(g) D.R. Wilson, R.D. Ernst, M.S. Kralik, Organometallics 3 (1984) 1442e1444;
(h) J.R. Bleeke, W.-J. Peng, Organometallics 5 (1986) 635e644;
(i) J.R. Bleeke, A.J. Donaldson, W.-J. Peng, Organometallics 7 (1988) 33e37.
[6] Cp*Co(C2H4)2/HBF4: (a) M. Brookhart, D.M. Lincoln, M.A. Bennett, S. Pelling,
J. Am. Chem. Soc. 112 (1990) 2691e2694;
[13] APEX II (Bruker, 2008) Bruker AXS Inc., Madison, Wisconsin, USA.
[14] SAINT (Bruker, 2008) Bruker AXS Inc., Madison, Wisconsin, USA.
[15] SADABS (Bruker, 2009) Bruker AXS Inc., Madison, Wisconsin, USA.
[16] (a) G.M. Sheldrick, Acta Cryst. A64 (2008) 112e122;
(b) G.M. Sheldrick, Acta Cryst. A71 (2015) 3e8;
(b) M. Brookhart, D.M. Lincoln, A.F. Volpe, G.F. Schmidt, Organometallics 8
(1989) 1212e1218.
(c) G.M. Sheldrick, Acta Cryst. C71 (2015) 3e8.
[7] An alternate synthetic method: G. Amandine, G. Lepesqueux, S. Sable,
S. Reymond, J. Cossy J. Org. Chem. 75 (2010) 5151e5163.
[17] POV-Ray 3.7 (www.povray.org).
[18] Diamond 3.2, Crystal Impact, Bonn, Germany (www.crystalimpact.com).
[19] C.F. Macrae, I.J. Bruno, J.A. Chisholm, P.R. Edgington, P. McCabe, E. Pidcock,
L. Rodriguez-Monge, R. Taylor, J. van de Streek, P.A. Wood, J. Appl. Cryst. 41
(2008) 466e470.
[8] A previous synthesis has been reported but was not successful in our hands
A. Eske, B. Goldfuss, A.G. Griesbeck, A. de Kiff, M. Kleczka, M. Leven,
€
J.M. Neudorfl, M. Vollmer, J. Org. Chem. 79 (2014) 1818e1829.
[9] M.J. Riveira, M.P. Mischne, Chem. Eur. J. 18 (2012) 2382e2388.
[10] From Method B: L.A. Carpino J. Org. Chem. 35 (1970) 3971e3972.
[20] A.L. Spek, Acta. Cryst. D65 (2009) 148e155.