Origin of the regioselectivity in an intramolecular nucleophilic addition to arene chromium tricarbonyl complexes

Article abstract of DOI:10.1016/j.ica.2010.07.037

The source of the regioselectivity in the intramolecular nucleophilic addition of nitrile-stabilized carbanions to arene chromium tricarbonyl complexes was investigated for seven different substitution patterns on the arenes. All of the arenes are 1,4-dio

Full text of DOI:10.1016/j.ica.2010.07.037

Inorganica Chimica Acta 364 (2010) 205–219
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
Origin of the regioselectivity in an intramolecular nucleophilic addition
to arene chromium tricarbonyl complexes
c
d
Steven Chamberlin a, Nilanjana Majumdar b, William D. Wulff b, , John V. Muntean , Robert L. Ostrander ,
⇑
Arnold L. Rheingold d
a Abbott Laboratories, Abbott Park, IL 60064, USA
b Department of Chemistry, Michigan State University, East Lansing, MI 48824, USA
c Chemistry Sciences and Engineering Division, Argonne National Laboratory, 9700 South Cass Avenue, Argonne, IL 60439, USA
d Department of Chemistry, University of California, San Diego, Urey Hall 5138, Mail Code 0358, 9500 Gilman Drive, La Jolla, CA 92093-0358, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Available online 4 August 2010
The source of the regioselectivity in the intramolecular nucleophilic addition of nitrile-stabilized carba-
nions to arene chromium tricarbonyl complexes was investigated for seven different substitution pat-
terns on the arenes. All of the arenes are 1,4-dioxygenated and the substitution varies in the oxygen
substituent and in the substituents of the arene carbons (hydrogen and alkyl). The regioselectivity is cor-
related with the preferred conformations of the chromium tricarbonyl group which in turn was deter-
mined by solution and solid-state 13C NMR spectroscopy, 1H NMR spectroscopy in solution as well as
X-ray diffraction. In the four complexes analyzed by X-ray diffraction and the three complexes analyzed
by solid-state 13C NMR spectroscopy, there was only one complex where it was found that the preferred
conformation of the –Cr(CO)3 is different in solution than it is in the solid-state.
Dedicated to Arnold L. Rheingold
Keywords:
Arene chromium tricarbonyl
Aromatic nucleophilic addition
X-ray diffraction
Solid-state 13C NMR spectroscopy
Solution 13C NMR spectroscopy
Regioselectivity
Ó 2010 Elsevier B.V. All rights reserved.
1. Introduction
of the phenol function in the phenol chromium tricarbonyl com-
plexes that are the primary products of the reactions of Fischer car-
Two of the most important synthetically significant chromium
(0)-mediated organic transformations were both first reported in
the literature in 1975 [1,2]. Both of these reactions made use of
organometallic compounds first synthesized in the laboratories of
Fischer [3,4]. In early 1975, the Semmelhack group described the
successful nucleophilic addition to benzene chromium tricarbonyl
with net substitution for hydride by a wide variety of carbanions
[1,5]. Latter the same year, Dötz reported that chromium Fischer
carbene complexes would react with alkynes to give arene chro-
mium tricarbonyl complexes in which the new arene ring was
the result of a formal [3+2+1] cycloaddition involving the assembly
of the alkyne, the carbene ligand and a carbon monoxide ligand
[2,6]. Since an arene chromium tricarbonyl complex is a product
of the reaction of a carbene complex with an alkyne, it seemed only
natural that these two reactions should be combined.
bene complexes with alkynes [8]. As indicated in Scheme 1, silyl
triflates are one of the electrophiles that can be used to trap the
phenol complexes to give hydroquinone silyl ether complexes of
the type 3 in an efficient manner and in good yields. Treatment
of these arene chromium tricarbonyl complexes with LDA leads
to deprotonation alpha to the nitrile and subsequent addition of
the resulting nitrile-stabilized carbanion to the coordinated ben-
zene ring at either of the two carbons originally derived from the
alkyne to give the
g
5-dienyl complexes 4 and 5. Oxidation of the
g
5-chromium tricarbonyl anionic intermediates 4 and 5 leads to
a mixture of the spirocyclic cyclohexadienone 6 and the tetrahy-
dronaphthalene 7. The distribution between these two products
depends on the nature of the substituents R1 and R2 and the tem-
perature of the reaction. In fact, in some cases the major product
obtained at room temperature differs from that obtained at
This did not happen until 1992 when our group investigated the
reaction of a number of chromium carbene complexes of the type 1
with 6-cyano-1-hexyne and other alkynes bearing carbanion stabi-
lizing groups [7]. The coupling of these reactions could not have
happened until methods were developed for the in situ protection
ꢀ78 °C indicating that the anionic
g
5-dienyl complexes 4 and 5
can interconvert and that the kinetic and thermodynamic products
are not the same.
We have previously reported the product distributions from the
intramolecular nucleophilic addition reactions of the six arene
chromium tricarbonyl complexes 8–14 and they are summarized
in Scheme 2 [5]. Treatment of the complex 8 with 1.2 equivalents
of LDA at ꢀ78 °C and then after 1.5 h quenching the reaction with
⇑
Corresponding author.
E-mail address: wulff@chemistry.msu.edu (W.D. Wulff).
0020-1693/$ - see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2010.07.037

206
S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
OSiR3
nucleophilic substitution include the electronic nature and substi-
CN
R3
R2
CN
OR1
tution pattern of ring substituents which can function as either
strong resonance donors or acceptors [9], avoidance of adverse ste-
ric interactions [10], maximization of favorable HOMO–LUMO
interactions [11], direction by total charge density [12] and activa-
tion of specific ring positions by a conformationally restricted
Cr(CO)3 fragment [13]. Complicating the picture even more is the
fact that several types of carbanions add to various arene–Cr(CO)3
complexes reversibly, so that products obtained can be the result
of either kinetically-controlled attack or of equilibration to a ther-
modynamically favored position [9h,14]. The emerging consensus
regarding the position of intermolecular kinetic attack by nucleo-
philes on Cr(CO)3-complexed arenes is to see the metal fragment
as exerting a major influence on the total charge distribution of
the complexed arene which directs attack to carbons eclipsed by
carbonyl ligands. Intramolecular nucleophilic additions of carba-
nions to arene chromium tricarbonyl complexes have not been
examined to as great an extent [9b,15].
1.5 equiv
2
(CO)5Cr
2,6-lutidine (2.5 equiv)
R3SiOTf (1.5 equiv)
CH2Cl2, 50 °C
R2
R3
(CO)3Cr OR1
1
3 (68 - 76%)
1) LDA (1.3 equiv)
THF
R3SiO
O
R3
R2
R3
R2
CN
CN
OR1
+
OR1
(CO)3Cr
6
4
2) I2
OSiR3
OSiR3
R3
R3
R2
As has been shown by the work of Jackson et al. [13a,b], Uemura
et al. [16], Ullenius and co-workers [9h] and Rose [13c], in systems
in which the –Cr(CO)3 fragment adopts a preferred configuration,
carbons eclipsed by a CO ligand are preferentially attacked. Though
still rotating, if the metal fragment adopted a preferred average po-
sition, it would lead to either the 1,3,5 or 2,4,6 carbons being rela-
tively electron poor and, hence, relatively activated for
nucleophilic attack. In our system this would translate to activa-
tion of either ipso or vicinal attack. ‘‘Thermodynamically-restricted
rotation”, as defined by Mislow and co-workers [17], has been used
to explain the behavior of substituted arene–Cr(CO)3 complexes in
the past. Unlike ‘‘kinetically-restricted rotation” [18] where con-
formers of the same relative energy are impeded from intercon-
verting by a high rotation barrier, ‘‘thermodynamically-restricted
rotation” is used to describe the phenomenon in which conformers
of different relative energy that are able to interconvert kinetically
are present in unequal amounts, with the lower energy conformer
predominating. In this case thermodynamically-restricted rotation
would be caused by the electron-releasing ability of the alkyl sub-
stituents to stabilize the relatively positively-charged eclipsed ring
carbons, inducing one of the two possible eclipsing conformations
of the metal fragment to be of lower energy.
The eclipsed conformations for the silyl-protected complexes
which differ only by the pattern of arene ring substitution are
shown in Scheme 3. Since these complexes are 1,4-dioxygenated
both the (1e) (carbons 1,3,5 eclipsed) and (4e) (carbon 2,4,6
eclipsed) conformations have one eclipsed carbon with an ether
substituent. For this analysis, the difference in electron-donating
ability between the methoxy and silyoxy groups is assumed to
be insignificant. For the purposes of this discussion the numbering
of the aromatic carbons in all of these complexes will be the same,
with the aromatic carbon bearing the methoxy substituent num-
bered as ‘‘1” and the aromatic carbon bearing the tethered nitrile
numbered ‘‘3”.
R2
OR1 CN
OR1 CN
(CO)3Cr
7
5
Scheme 1.
iodine at ꢀ78 °C gave to the spirocyclic dienone product 15 result-
ing from ipso-addition in 73% yield. On the other hand, if the reac-
tion temperature was allowed to rise to 0 °C after the addition of
LDA and the reaction quenched with iodine after an hour at 0 °C,
the dienone 15 was not detected and instead the fused tetralin
16 resulting from vicinal addition was isolated in 67% yield. This
was interpreted to mean that complex 4 (Scheme 1) is the kinetic
g
5-dienyl complex and that it is completely converted to the ther-
modynamic
to 0 °C.
g
5-dienyl complexes 5 when the temperature is raised
The kinetic and thermodynamic products are less clear-cut for
many of the other arene complexes. For the TBS substituted com-
plex 9, the spirocyclic product 17 was also formed in good yields
at ꢀ78 °C; however, when the temperature was raised to 0 °C,
the 1H NMR spectrum of the reaction mixture indicated the ab-
sence of both of the products 17 and 18 and the presence of two
other compounds (ꢁ30% each), one of which was identified as
the metal free arene derived from complex 9. Interestingly, while
the TBS complex 9 gave the spirocyclic product 17 at ꢀ78 °C, the
TIPS complex 10 gave the isomeric vicinal substituted product 18
at ꢀ78 °C. The 6-methyl substituted arene complex 11 displayed
a different behavior, giving the vicinal substituted product 20
whether the intramolecular nucleophilic addition was carried out
at ꢀ78 °C or at 0 °C. It was thought the regiochemistry could be re-
versed for steric reasons by employing cyclohexyl substituted
arene complex 12, but instead the reaction gave dramatically re-
duced yields of the same regioisomer. The regioselectivity of the
6-methyl complex 11 was reversed at ꢀ78 °C for the 5-methyl
complex 13 which gave the spirocyclic product 21 in 73% yield.
Raising the temperature to 0 °C led to a switch in preference for
the vicinal substituted product 22a but only in very low yield. Un-
like, the reactions of the arene complexes 9 and 10, there was no
significant effect of TBS versus TIPS groups in the complexes 13
and 14 since both led to the spirocyclic product 21 at ꢀ78 °C.
The reasons for the regioselectivity outcomes for the reactions in
Scheme 2 are not entirely readily evident and is the subject of
the present work.
In looking at the alkyl substituents, the conformations having
more eclipsed carbons with alkyl, as opposed to hydrogen substit-
uents, would be expected to be electronically more stable. This
would predict conformations 9 (1e) would be preferred over 9
(4e), 13 (1e) would be preferred over 13 (4e), and 8 (1e) would
be preferred over 8 (4e). In each of these cases the kinetic product
of nucleophilic attack is the result of attack at C-3, or attack at the
ring position which by this analysis is activated by being eclipsed
with a CO ligand. Because of the para disposition of the ether
and alkyl substituents, a prediction for complex 11 cannot be
made.
The regioselectivity seen in the addition of nucleophiles to
substituted benzene chromium tricarbonyl has been extensively
investigated in the past [5]. Factors which effect the position of
The goal of the present work is to interpret the regioselectivities
observed in the reactions outlined in Scheme 2 in terms of the
conformation of the chromium tricarbonyl groups in the arene

S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
207
OTBS
O
OTBS
1) LDA,
CN
-78 °C
2) X °C
3) I2
CN
(CO)3Cr OMe
OMe
OMe CN
8
16
X °C
15
-78
°C
73%
< 5%
< 5%
67%
0 °C
OR
O
OH
1) LDA,
-78 °C
CN
CN
CN
CN
2) X °C
3) I2
(CO)3Cr OMe
OMe
OMe CN
R
TBS
TBS
TIPS
X °C
-78 °C
0 °C
17
R = TBS
R = TIPS
9
10
18
72%
< 5%
< 5%
< 5%
< 5%
62%
°C
-78
OTBS
O
OTBS
1) LDA,
-78 °C
2) X °C
3) I2
CN
(CO)3Cr OR
OR
OR CN
20a,b
62%
52%
18%
R = Me
R = C6H11
11
12
R
Me
Me
X °C
19a,b
< 5%
< 5%
< 5%
-78
°C
0 °C
-78 °C
C6H11
OR
O
OR
1) LDA,
-78
°C
2) X °C
3) I2
CN
(CO)3Cr OMe
OMe
OMe CN
R
X °C
-78 °C
0 °C
21
R = TBS
R = TIPS
22a,b
< 5%
13%
13
14
TBS
TBS
TIPS
73%
< 5%
48%
-78 °C
< 5%
Scheme 2.
complexes 8–14. The products in Scheme 2 obtained at ꢀ78 °C are
considered kinetic addition products while those obtained at 0 °C
are considered thermodynamic products. Kundig et al. has found
that intermolecular addition reactions conducted and quenched
below ꢀ50 °C, or which use HMPA as cosolvent, give kinetic addi-
tion products [14b]. The role of the HMPA is to coordinate to the
cation and decrease the interaction between the cation and the
carbanions-stabilizing group which is necessary for reversible
addition. We did not perform the reactions in the presence of
HMPA. In the instance of complex 8, we did get a clean rearrange-
ment to a thermodynamic product. Warming the reactions of com-
plex 9 led to mixtures from which no addition products could be
obtained. With complex 11 warming the reaction afforded the
same major product, meaning either the kinetic and thermody-
namic products are the same, as was found to be the case with ani-
sole-Cr(CO)3 and the anion of isobutyronitrile [14b] or there is a
rapid conversion of a kinetic product to the thermodynamic addi-
tion product 20a [9h]. Because of our results with complex 8 we
consider the former possibility more likely. In no instance did we
see any signs of formation of spirocycle 19. Complex 13 also rear-
ranged to a different product upon warming, although not as
cleanly as complex 8.
The conformations of the Cr(CO)3 group in each complex have
been determined by a combination of 13C NMR spectroscopy in
solution and X-ray structure determination in the solid-state and
by the use of 13C NMR spectroscopy in the solid-state to correlate
the solution state with the solid-state. It is not often that solid-
state 13C NMR spectroscopy has been used to correlate solution
and solid-state structures in arene chromium tricarbonyl com-
plexes [18a] and to our knowledge it has not been used, in turn,
to interpret regioselectivities in nucleophilic addition to these
complexes. Solid-state X-ray structures could be determined for
four complexes (9–11 and 14) and solid-state 13C NMR spectra
were obtained for three complexes (9, 11 and 14). To determine
the conformations of the Cr(CO)3 in the various complexes by 13C
NMR spectroscopy it was necessary to assign all six of the arene
carbon atoms in each complex (8–14, Scheme 2) and also to assign
all six of the arene carbon atoms in each of the corresponding free
arenes (8a–14a, Scheme 3). All of the free and complexed arenes
examined in this work have six distinct aromatic resonances. The

208
S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
OR
OTBS
OR
3
OTBS
4
4
4
4
3
3
3
CN
CN
CN
CN
5
5
5
5
2
2
2
2
6
6
6
6
1
1
1
1
OMe
OMe
OMe
OR
9a
10a
8a
13a
14a
R = TBS
R = TIPS
11a
12a
R = TBS
R = TIPS
R = Me
R = C6H11
OTBS
OTBS
OTBS
OTBS
OMe
CN
CN
CN
CN
CN
CN
CN
OMe
OMe
OMe
9 (1e)
8 (1e)
11 (1e)
13 (1e)
OTBS
OTBS
OTBS
OTBS
CN
OMe
8 (4e)
OMe
OMe
OMe
9 (4e)
11 (4e)
13 (4e)
Scheme 3.
assignment of each arene carbon in the Cr(CO)3 complexes and the
free arenes was done by a combination of introducing a 13C label at
position 4 and introducing a deuterium at position 2. The former
was done be preparing carbene complexes of the type 1 (Scheme 1)
with 13C enriched carbon monoxide ligands. The benzannulation
reaction with alkyne 2 gave arene complexes enriched in 13C at
carbon 4, since this carbon is derived from a carbon monoxide li-
gand [6]. Carbons C-3 and C-5 can be identified by 13C–13C coupling
to C-4. Although when C-5 bears an alkyl substituent, C-3 and C-5
cannot be distinguished in this way, since the –Cr(CO)3 fragment
has C3v symmetry, for this analysis, they do not need to be further
assigned. Introduction of a deuteron at the 2-position was accom-
plished by performing the benzannulation reaction with alkyne 2
in which the alkenyl hydrogen had been replaced by deuterium
and allowed the assignment of C-2 and C-6.
data reported in this paper were obtained nearly two decades ago
using scintillation detectors. Various methods not applicable to
modern CCD-detector data collections were employed to control
the time required to obtain adequate quality data, including
restrictions in 2h to allow only the collection of observed data,
minimal redundancy and variable-speed scan times. Alerts found
in CHECKCIF output that would be considered serious today were
necessary when these data were obtained. Nonetheless, all data re-
ported in this paper are considered fully reliable by modern
criteria.
2.2. Synthesis of arene chromium tricarbonyl complex 8 and the free
arene 8a
Carbene complex 1a [19] (R1 = Me, R2 = H, R3 = TBS) (0.3283 g,
0.872 mmol) and alkyne 2 (0.190 g, 1.77 mmol) were combined
in 18 mL CH2Cl2 in a 50 mL flame dried pear-shaped single-necked
flask in which the 14/20 joint was replaced by a high vacuum T-
shaped Teflon value. The system was deoxygenated by the
freeze–thaw method and after the third cycle, the flask was
back-filled with argon at room temperature. The flask was sealed
by closing the Teflon value and the flask was then heated at
65 °C for 11 h. The temperature of the oil bath was then raised to
73 °C and the reaction mixture was heated until it was clear and
yellow (4 h). After removal of the solvents, the crude product
was purified by chromatography on silica gel with 1:1:6 mixture
of ether/methylene chloride/hexane to give arene–Cr(CO)3 com-
plex 8 (0.269 g, 0.590 mmol, 68%) as a yellow solid. Spectral data
for 8: (yellow prisms [ether/pentane] mp 84–85 °C, Rf = 0.24,
1:1:4) 1H NMR (CDCl3) d 0.32 (s, 3 H), 0.36 (s, 3 H), 0.98 (s, 9 H),
1.70–1.83 (m, 4 H), 2.32–2.38 (m, 1 H), 2.41 (t, 2 H, J = 6.4 Hz),
2.80–2.88 (m, 1 H), 3.63 (s, 3 H), 5.00 (dd, 1 H, J = 7.0, 2.6 Hz),
5.16 (d, 1 H, J = 7.0 Hz), 5.20 (d, 1 H, J = 2.6 Hz); 13C NMR (CDCl3)
d ꢀ4.8, ꢀ4.5, 16.9, 17.9, 25.1, 25.3, 29.2, 29.8, 55.9, 75.8 (C-6),
82.1 (C-5), 83.2 (C-2), 104.3 (C-3), 119.2 (-CN), 130.0 (C-4), 137.6
(C-1), 234.1; IR (neat film) 2932w, 2250vw, 1955vs, 1866vs,
1472m, 1258m, 878w, 672w cmꢀ1; mass spectrum m/z (rel. inten-
sity) 455 (M+, 5), 371 (95), 356 (5), 314 (15), 262 (100). Anal. Calc.
for C21H29CrNO5Si: C, 55.37; H, 6.42; N, 3.07. Found: C, 55.50; H,
6.35; N, 2.89%.
2. Experimental
2.1. General procedures, methods and materials
Flash column chromatography was performed on EM Science
330–400 mesh silica gel. Hexanes (from bulk) were distilled before
use, while other solvents (HPLC grade) were used without further
purification. Solvent systems were either ternary mixtures of
diethyl ether, dichloromethane, and hexanes, which are referred
to in this order (e.g., 1:1:4, diethyl ether, dichloromethane, and
hexanes, v/v/v), mixtures of ethyl acetate and hexanes (v/v), or
hexanes. Analytical TLC was performed on 3.5 ꢂ 7.5 cm Baker-flex
IB2-F plastic plates with a mm silica gel layer. Low resolution mass
spectra were recorded on a Finnigan 1015 instrument. High resolu-
tion mass spectra were recorded on a VG analytical 7070E mass
spectrometer. Elemental analyses were performed by Galbraith
Laboratories, Inc., Knoxville TN. Solid-state 13C NMR spectra were
recorded on a Bruker 300 AM instrument at Argonne National Lab-
oratory with an external reference of tetrakis-trimethylsilylsilane.
Single crystals of compounds 11, 9, and 10 were grown at ambi-
ent temperatures from 50% saturated ether solutions layered with
an equal volume of pentane. Single crystals of 14 were grown at
ambient temperatures from a saturated methylene chloride solu-
tion layered with twice the volume of pentane. All crystallographic

S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
209
Table 1
2.2.1. Assignment of the arene carbons of complex 8
Variable temperature 13C NMR data for complex 8 (CD2Cl2).
Heating a CH2Cl2 solution of carbene complex 1a at 65 °C for
24 h under 1 atmosphere of 13CO led to complete decomposition
of the carbene complex. Thus the carbon resonances were only ten-
tatively assigned with the aid of analogy to the arene–Cr(CO)3
complexes in which C-4 had been labeled. Assignment of the C-1,
C-4, and C-3 resonances were straightforward. In all cases the res-
onance C-1 (C-OMe) is farther downfield than the one for C-4. The
resonance for C-3 showed no NOE enhancement and is in the range
characteristic of alkyl-substituted aromatic carbons of arene–
Cr(CO)3 complexes. Assignment of the Car-H resonances was less
simple. The farthest upfield Car-H resonance, 75.8 ppm, was as-
signed to C-6, since this carbon in the free arene was calculated
to have the highest upfield shift, situated para to an alkyl substitu-
Temp (°C)
CO
C-1
C-4
C-3
C-2
C-5
C-6
20
0
ꢀ25
ꢀ50
ꢀ60
234.22
234.19
234.11
234.00
234.00
137.60
137.54
137.44
–
130.01
129.72
129.29
128.87
128.71
104.62
104.67
104.73
104.80
104.83
83.41
83.37
83.29
83.19
83.14
82.23
82.13
82.03
82.00
82.00
75.92
75.61
75.17
74.69
74.51
137.26
1 H, J = 8.6 Hz); 13C NMR (CDCl3) d ꢀ4.2, 17.0, 18.1, 25.0, 25.8,
28.9, 29.8, 55.5, 111.5 (C-6), 115.7 (C-5), 118.9 (C-2), 119.5 (–
CN), 132.6 (C-3), 147.3 (C-4), 153.7 (C-1); IR (neat film) 2248vw,
1498s, 1254m, 1222s cmꢀ1; mass spectrum m/z (rel. intensity)
319 (M+, 95), 304 (40), 293 (10), 276 (5), 263 (100), 247 (15),
234 (90), 220 (45), 207 (40), 183 (95), 165 (80). HRMS calc for
ent. This would mean a
Dd on complexation of 35.7 ppm. It was
considered unlikely that the carbon resonance at 111.5 ppm in
the free arene would correspond to either of the other Car-H reso-
C
18H29NO2Si m/z 319.1968, meas 319.2003. HRMS on M+–CH3 calc
for C17H26NO2Si m/z 304.1733, meas 304.1758.
nances (82.1, 83.2 ppm), since this would lead to a
Dd value of
The aromatic carbon resonances were assigned through correla-
tion with shifts calculated using additive substituent effects on car-
bons of mono-substituted benzenes [20] substituting the values of
ethyl and methoxyl for the C-3 tether and the TBS-oxy group,
respectively. The calculated values (observed resonance) are as fol-
lows: C-1, 152.2 (153.7); C-4, 151.7 (147.3); C-3, 130.7 (132.6); C-
5, 115.1 (115.7), C-2, 114.6 (118.9), C-6, 112.5 (111.5). The ratio-
nale for assigning the C-2 and C-5 resonances is included in the dis-
cussion above of the assignments of the aromatic resonances of
arene–Cr(CO)3 complex 8.
either 29.4 or 28.3 ppm, values well outside the range (32.8–
35.1 ppm) seen for the other Car-H
Dd values of the fully assigned
complexes 9–14.
The tentative assignment of the C-2 and C-5 resonances is based
on the following analysis. The resonances of arene–Cr(CO)3 com-
plex 8 at 82.1 and 83.2 ppm correspond to the resonances of free
arene 8a at 115.7 and 118.9. The resonance at 82.1 ppm is assigned
to correspond to that at 115.7 ppm, while that at 83.2 ppm corre-
sponds to that at 118.9 ppm, since this gives
and 35.7 ppm. The opposite correlation would give
D
d values of 33.6
d values of
D
36.8 and 32.5 ppm. The first value is well outside the established
range (32.8–35.1 ppm). The assignment of C-2 and C-5 then simpli-
2.3. Synthesis of arene chromium tricarbonyl complex 9 and the free
arene 9a
fies to which pair of resonances, [82.1/115.7,
Dd = 33.6 ppm] and
[83.2/118.9, d = 35.7 ppm], belongs to C-2 and which to C-5. Sim-
D
A solution of carbene complex 1b (R1 = Me, R2,R3 = –(CH2)4–)
[21] (0.1213 g, 0.3839 mmol), alkyne 2 (0.062 g, 0.58 mmol), 2,6-
lutidine (0.11 mL, 0.96 mmol), and TBSOTf (0.13 mL, 0.58 mmol)
in 7.7 mL of CH2Cl2 was added to a 50 mL flame dried pear-shaped
single-necked flask in which the 14/20 joint was replaced by a high
vacuum T-shaped Teflon value. The system was deoxygenated by
the freeze–thaw method and after the third cycle, the flask was
back-filled with argon at room temperature. The flask was sealed
by closing the Teflon value and the flask was then heated at
50 °C under argon for 23 h. After removal of the solvents, the crude
product was purified by chromatography on silica gel with 1:1:4
mixture of ether/methylene chloride/hexane to give the arene–
Cr(CO)3 complex 9 (0.1403 g, 0.275 mmol, 72%). Spectral data for
9: (yellow needles [ether], mp 119–120 °C, Rf = 0.30, 1:1:4) 1H
NMR (CDCl3) d 0.37 (s, 3 H), 0.39 (s, 3 H), 1.01 (s, 9 H), 1.49–1.65
(m, 1 H), 1.69–1.82 (m, 6 H), 1.91–1.98 (m, 1 H), 2.28–2.32 (m, 1
H), 2.41 (t, 2 H, J = 7.0 Hz), 2.49–2.54 (m, 1 H), 2.69–2.80 (m, 4
H), 3.71 (s, 3 H), 4.86 (s, 1 H); 13C NMR (CDCl3) d ꢀ3.24, ꢀ1.96,
17.18, 18.74, 21.08, 21.83, 23.71, 25.09, 25.15, 25.93, 29.55,
29.74, 55.90, 73.42 (C-2, C-ar-H), 98.87 (C-6), 102.37 (C-3),
106.09 (C-5), 119.19 (CN), 125.61 (C-4), 137.27 (C-1, C-ar-OMe),
234.97; IR (neat film) 2247vw, 1946s, 1857s, 1457m, 1429m,
1259m, 1091m cmꢀ1; mass spectrum m/z (rel. intensity) 509 (M+,
10), 425 (45), 41 (10), 373 (15), 316 (100), 288 (10). Anal. Calc.
for C25H35CrNO5Si: C, 58.92; H, 6.92; N, 2.75. Found: C, 58.77; H,
7.00; N, 2.53%.
ple calculations based on substituent effects [20] on the uncom-
plexed arene 8a predicts the C-2 resonance at 114.6 ppm and the
C-5 resonance at 115.1 ppm due to a slight upfield shift of C-2 ver-
sus C-5 since it is ortho to a straight-chain alkyl substituent. This
would suggest assigning the more upfield resonances to C-2. How-
ever, C-2 and C-6 are related by symmetry to the C3v symmetric
metal fragment; the largest difference in a single complex for the
C-2 and C-6
and 33.1 for 14). For 13 the difference in
C-6 is 0.5 ppm. If the assignment for C-2 is the resonances at
82.1 ppm (8) and 115.7 ppm (8a), the difference in the d values
for C-2 and C-6 would be 2.1 ppm. Assignment of C-2 as the other
pair of resonances gives the same value of d for both C-2 and C-6.
D
d values is 1.4 ppm (
D
d upon complexation = 34.5
D
d values for C-2 and
D
D
The assignments of C-2, C-5, and C-6 are reasonable when com-
pared to similar carbons in the fully assigned complexes, especially
when looking at the values of
Dd upon complexation. The Dd value
for both C-2 and C-6 (35.7 ppm) is consistent with the values ob-
tained for those carbons thought to be on-average staggered by
the –Cr(CO)3 fragment. For complexes 13 and 14, the
C-2/6 are 35.1/34.6 and 34.5/33.1 ppm, respectively. In complex
9 the d value for C-2 is 34.4 ppm, while in complex 11 it is
34.7 ppm for C-5. The d value for C-5 (33.6 ppm) is also consis-
Dd values
D
D
tent with those of on-average eclipsed carbons (C-2 of 10,
32.3 ppm; C-2 of 11, 32.8 ppm). The 13C NMR spectrum of complex
8 was also investigated at a number of temperatures down to
ꢀ60 °C in CD2Cl2 and the data are presented in Table 1.
2.2.2. Synthesis of 8a and the assignment of the arene carbons
The arene 8a was isolated as a minor component of a sequential
benzannulation/nucleophilic addition reaction (yield not deter-
mined). Spectral data for 8a: (colorless oil, Rf = 0.42, 20% EtOAc/
hexanes) 1H NMR (CDCl3) d 0.23 (s, 6 H), 1.02 (s, 9 H), 1.68–1.78
(m, 4 H), 2.34 (t, 2 H, J = 6.9 Hz), 2.60 (t, 2 H, J = 7.2 Hz), 3.75 (s, 3
H), 6.60 (dd, 1 H, J = 8.6, 3.0 Hz), 6.65 (d, 1 H, J = 2.8 Hz), 6.68 (d,
2.3.1. Assignment of the arene carbons of complex 9
The C-4 13C, 2H-labeled arene–Cr(CO)3 complex 9 was prepared
by stirring carbene complex 1b in a minimal amount of hexanes for
24 h at 50 °C under 1 atmosphere of 99% 13CO. The benzannulation
reaction was performed as described above for unlabeled complex
9 except that instead of blanketing the reaction with argon, after
the final freeze–pump–thaw cycle the reaction vessel was charged

210
S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
with 1 atmosphere of 13C-labeled CO. In addition the alkenyl pro-
ton of the alkyne 2 had been replaced with a deuteron. After
25 h at 55 °C the labeled complex 9 was obtained in 57%. The signal
at d 125.61 was significantly enhanced and was assigned as the
resonance for Car -OTBS (C-4). The signal at 137.27 was assigned
as C-ar-OMe (C-1). The signal at 73.42 was assigned as that for
C-2 and this was confirmed when in the 2H labeled complex the
intensity of the signal at 73.42 was nearly eliminated. This was fur-
ther confirmed by a gHMQC experiment in which coupling was
established between the proton singlet at 4.86 and the carbon at
73.4 The aromatic signals at 106.09 and 102.37 were surrounded
by doublets (J = 57 and 58 Hz, respectively) resulting from 13C to
13C coupling, indicating these were the resonances for C-3 and C-
5. The signal at 98.87 was free of any such doublet and was as-
signed as C-6. The carbons C-3 and C-5 were distinguished by
the following experiments. A TOCSY 1D experiment involving irra-
diation at the triplet at 2.41 (next to nitrile) identified all of the
protons in the four carbon aliphatic chain. One of these protons
was not overlapped with the others (m, 2.28–2.32) and a gHMBC
experiment revealed that this proton is coupled to C-3 but not C-
5. The 13C NMR spectrum of complex 9 was also investigated at a
number of temperatures down to ꢀ60 °C in CD2Cl2 and the data
are presented in Table 2. Six signals were identified in the aromatic
region of the solid-state 13C NMR spectrum that correlated well
with the chemical shifts in solution. The assignments made for
these carbons are: C-1 (138.6), C-4 (125.4), C-5 (107.3), C-3
(105.2), C-6 (101.1), C-2 (75.7).
plet at 2.59–2.64. In a gHMBC experiment it was shown that this
proton is coupled to C3 and the nitrile carbon.
2.4. Synthesis of arene chromium tricarbonyl complex 10 and the free
arene 10a
A solution of carbene complex 1b (R1 = Me, R2, R3 = –(CH2)4–)
[21] (0.204 g, 0.647 mmol), alkyne 2 (0.133 g, 1.24 mmol), 2,6-luti-
dine (0.19 mL, 1.62 mmol), and TIPSOTf (0.26 mL, 0.97 mmol) in
13 mL of CH2Cl2 was added to a 50 mL flame dried pear-shaped
single-necked flask in which the 14/20 joint was replaced by a high
vacuum T-shaped Teflon value. The system was deoxygenated by
the freeze–thaw method and after the third cycle, the flask was
back-filled with argon at room temperature. The flask was sealed
by closing the Teflon value and the flask was then heated at
50 °C under argon for 23 h. After removal of the solvents, the crude
product was purified by chromatography on silica gel with 1:1:4
mixture of ether/methylene chloride/hexane to give after crystalli-
zation the arene–Cr(CO)3 complex 10 (0.257 g, 0.466 mmol, 72%).
Spectral data for 10: (fine yellow needles [ether/pentane], mp
135–136 °C, Rf = 0.28, 1:1:4) 1H NMR (CDCl3) d 1.15 (d, 18 H,
J = 6.8 Hz), 1.22–1.30 (m, 3 H), 1.55–1.68 (m, 1 H), 1.78–1.98 (m,
7 H), 2.44–2.51 (m, 2 H), 2.57–2.67 (m, 4 H), 2.73–2.78 (m, 1 H),
2.92 (dt, 1 H, J = 16.4, 5.4 Hz), 3.67 (s, 3 H), 5.11 (s, 1 H); 13C
NMR (CDCl3) d 14.4, 17.1, 18.03, 18.06, 21.3, 21.6, 23.0, 25.61,
25.65, 28.3, 29.0, 56.3, 75.5 (C-2, Car-H), 99.4 (C-3/5), 101.1 (C-
6), 103.1 (C-3/5), 119.3 (CN), 132.4 (C-4), 134.2 (C-1, C-OMe),
235.0; IR (neat film) 2248w, 1943vs, 1901m, 1862vs, 1453m,
1422m cmꢀ1; mass spectrum m/z (rel. intensity) 551 (M+, 22),
467 (100), 415 (45), 372 (85), 340 (15). Anal. Calc. for C28H41CrNO5-
Si: C, 60.96; H, 7.49; N, 2.54. Found: C, 60.62; H, 7.53; N, 2.23%.
2.3.2. Synthesis of 9a and the assignment of the arene carbons
Iodine (5 equiv) was added as a solid to a stirred solution of la-
beled arene–Cr(CO)3 complex 9 in 10 mL THF under air at rt. TLC
showed a clean conversion to the desired free arene. After 1 h,
the mixture was washed with 10% Na2S2O3 and dried with brine
and MgSO4. Chromatography (20% EtOAc/hexanes) through a short
silica gel column afforded arene 9a. Spectral data for 9a: (pale yel-
low solid, mp 57–60 °C, Rf = 0.48, 20% EtOAc/hexanes); 1H NMR
(CDCl3) d 0.19 (s, 6 H), 1.02 (s, 9 H), 1.65–1.72 (m, 8 H), 2.33 (t, 2
H, J = 6.9 Hz), 2.59–2.64 (m, 6 H), 3.77 (s, 3 H), 6.41 (s, 1 H); 13C
NMR (CDCl3) d ꢀ2.83, 17.07, 18.81, 22.35, 22.60, 23.30, 25.02,
25.62, 26.20, 29.37, 29.98, 55.45, 107.93 (C-2, C ar-H), 119.72
(CN), 124.73 (C-6), 127.73 (C-5), 129.90 (C-3), 144.80 (C-4),
151.74 (C-1, C-OMe); IR (neat film) 2246vw, 1464m, 1414w,
1230s, 1093m, 883m cmꢀ1; mass spectrum m/z (rel. intensity)
373 (M+, 55), 316 (M-t-Bu, 100), 288 (10), 273 (5), 253 (5), 235
(5), 219 (5), 199 (10), 185 (5); HRMS calc for C22H35NO2Si m/z
373.2437, meas 373.2453.
2.4.1. Assignment of the arene carbons of complex 10
The 13C, 2H-labeled arene–Cr(CO)3 complex 10 was prepared by
stirring carbene complex 1b in a minimal amount of hexanes for
24 h at 50 °C under 1 atmosphere of 99% 13C-labeled CO. The benz-
annulation reaction was conducted as above, except that instead of
blanketing the reaction with argon, after the final freeze–pump–
thaw cycle the reaction vessel was charged with 1 atmosphere of
99% 13C-labeled CO. In addition the alkenyl proton had been re-
placed with a deuteron. After 25 h at 55 °C the 13C, 2H-labeled 10
was obtained in 56% yield. The signal at d 132.4 was significantly
enhanced and was assigned as the resonance for Car-OTBS (C-4).
The signal at 134.2 was assigned as Car-OMe (C-1). The signal at
75.5 had already been assigned as that for C-2 on the basis of its
greater intensity in the spectrum of the unlabeled complex. This
assignment was confirmed when in the 2H labeled complex the
intensity of the signal at 75.5 was nearly eliminated. The aromatic
signals at 103.1 and 99.4 were surrounded by doublets (J = 54 and
60 Hz, respectively) resulting from 13C to 13C coupling, indicating
these were the resonances for C-3 and C-5. The signal at 101.1
was free of any such doublet and was assigned as C-6. The 13C
NMR spectrum of complex 10 was also investigated at a number
of temperatures down to ꢀ60 °C in CD2Cl2 and the data are pre-
sented in Table 3.
The aromatic carbon signals were assigned on the basis of the
following evidence: d 144.80 (C-OTBS, increased signal intensity
due to 13C label), 151.74 (C-OMe, lack of increased signal intensity),
127.73 and 129.90 (C-3 and C-5, surrounded by 13C–13C doublets,
J = 69 and 68 Hz, respectively), 107.93 (Car-H, decrease in signal
intensity in 2H-labeled compound), 124.73 (C-6, lack of 13C–13
C
doublets). Carbons C-3 and C-5 were distinguished by the follow-
ing experiments. A gCOSY experiment reveals that the triplet at
2.33 next to the nitrile is 3-bond coupled to a proton in the multi-
2.4.2. Synthesis of 10a and the assignment of the arene carbons
Iodine (0.36 g, 1.42 mmol, 7 equiv) was added as a solid to a
solution of labeled arene–Cr(CO)3 complex 10 in 3 mL THF stirred
in air at rt. TLC showed formation of both the desired free arene
and the desilylated uncomplexed phenol. After 45 min, the mixture
was washed with 10% Na2S2O3 and dried with brine and MgSO4.
Chromatography on silica gel (20% EtOAc/hexanes) afforded 13C,
2H-labeled arene 10a (yield not determined). Spectral data for
10a: (colorless oil, Rf = 0.54, 20% EtOAc/hexanes) 1H NMR (CDCl3)
d 1.07 (s, 18 H), 1.25–1.35 (m, 3 H), 1.71–1.76 (m, 8 H), 2.35 (t, 2
Table 2
Variable temperature 13C NMR data for complex 9 (CD2Cl2).
Temp (°C) CO
C-1
C-4
C-5
C-3
C-6
C-2
20
0
ꢀ25
ꢀ50
ꢀ60
235.01 137.23 125.44 106.35 102.89 98.71 73.71
234.96 137.27 124.99 106.46 102.97 98.48 73.48
234.91 137.29 124.31 106.56 103.05
234.81 137.29 123.66 106.64 103.14 97.78 72.90
234.78 137.27 123.40 106.66 103.15 97.67 72.80
–
73.16

S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
211
13CO and sealed. After 24 h at 50 °C normal workup and chroma-
tography (20% EtOAc/hexanes) afforded the labeled complex 11
(yield not determined). The aromatic carbon resonances were as-
signed on the basis of the following evidence: 131.4 (C-4, C-OTBS,
increased signal intensity), 134.4 (C-1, C-OMe, lack of increased
signal intensity), 100.3 (C-3, surrounded by a 13C–13C doublet,
J = 59 Hz), 86.3 (C-5, surrounded by a 13C–13C doublet, J = 53 Hz)
79.3 (C-2, Car-H, decrease in signal intensity in 2H-labeled com-
pound), 97.8 (C-6, lack of 13C–13C doublet). The 13C NMR spectrum
of complex 11 was also investigated at a number of temperatures
down to ꢀ60 °C in CD2Cl2 and the data are presented in Table 4.
Six signals were identified in the aromatic region of the solid-state
13C NMR spectrum that correlated well with the chemical shifts in
solution. The assignments made for these carbons are: C-1 (137.7),
C-4 (127.2), C-3 (108.1), C-6 (94.5), C-5 (90.5), C-2 (78.4).
Table 3
Variable temperature 13C NMR data for complex 10 (CD2Cl2).
Temp (°C)
CO
C-1
C-4
C-3/5
C-6
C-3/5
C-2
23
0
235.7
235.6
235.5
235.4
235.4
235.3
235.3
134.9
134.7
134.4
134.3
134.2
134.1
134.1
133.0
132.8
132.5
132.3
132.1
132.0
132.0
103.9
103.9
104.0
104.0
104.1
104.1
104.1
101.6
101.4
101.0
100.7
100.5
100.3
100.3
100.6
100.4
100.4
100.3
100.2
100.2
100.2
76.5
76.2
75.7
75.5
75.3
75.1
75.1
ꢀ40
ꢀ60
ꢀ78
ꢀ90
ꢀ95
H, J = 6.8 Hz), 2.61–2.67 (m, 6 H), 3.76 (s, 3 H), 6.39 (s, 1 H); 13C
NMR (CDCl3) d 12.3, 17.5, 17.7, 22.3, 22.6, 23.4, 25.3, 25.4, 29.6,
30.3, 55.4, 107.8 (C-2, Car-H), 119.7 (CN), 124.6 (C-6), 127.3 (C-3/
5), 129.3 (C-3/5), 146.2 (C-4), 151.5 (C-1, C-OMe); IR (neat film)
2246vw, 1464m, 1230m, 883m cmꢀ1; mass spectrum m/z (rel.
intensity) 415 (M+, 90), 372 (M–i-Pr, 100), 344 (10), 330 (5), 316
(10), 302 (5), 289 (4), 275 (3), 260 (5), 247 (5), 233 (5), 217 (5),
2.5.2. Synthesis of 11a and the assignment of the arene carbons
Iodine (0.40 g, 1.57 mmol, 5 equiv) was added as a solid to a
solution of arene–Cr(CO)3 complex 11 (0.147 g, 0.313 mmol) stir-
red in 2 mL THF under air. TLC indicated 11 had been consumed
in less than 10 min. The mixture was diluted with ether, washed
with 10% Na2S2O3, 1% HCl, and brine, then dried over MgSO4. Filtra-
tion through a short plug of 1:1 silica gel/Celite and removal of sol-
vent afforded arene 11a (0.084 g, 0.252 mmol, 80%) which was
characterized without further purification. Spectral data for 11a:
(colorless oil, Rf = 0.49, 1:1:4) 1H NMR (CDCl3) d 0.22 (s, 6 H),
1.02 (s, 9 H), 1.68–1.72 (m, 4 H), 2.15 (s, 3 H), 2.34 (t, 2 H,
J = 6.8 Hz), 2.59 (t, 2 H, J = 7.2 Hz), 3.77 (s, 3 H), 6.55 (s, 2 H); 13C
NMR (CDCl3) d ꢀ4.2, 15.9, 17.0, 18.1, 25.0, 25.8, 29.2, 29.7, 55.9,
112.1 (C-2, Car-H), 119.7 (CN), 121.0 (C-5), 124.9 (C-6), 129.0 (C-
3), 146.6 (C-4), 151.9 (C-1, C-OMe); IR (neat film) 2246vw,
1504m, 1464m, 1398m, 1259m, 1212s, 1030m cmꢀ1; mass spec-
trum m/z (relative intensity) 333 (M+, 100), 318 (MꢀCH3, 30),
277 (100), 248 (85), 233 (50), 219 (90), 207 (70), 195 (90), 183
(100), 163 (25), 151 (85), 138 (40), 124 (85); HRMS calc for
186 (15); HRMS calc for
415.2935.
C25H41NO2Si m/z 415.2907, meas
The aromatic carbon signals were assigned on the basis of the
following evidence: d 146.2 (C-OTBS, increased signal intensity),
151.6 (C-OMe, lack of increased signal intensity) 127.3 and 129.3
(C-3 and C-5, surrounded by 13C–13C doublets, J = 71 and 76 Hz,
respectively), 107.8 (Car-H, decrease in signal intensity in 2H-la-
beled compound), 124.6 (C-6, lack of 13C–13C doublets).
2.5. Synthesis of arene chromium tricarbonyl complex 11 and the free
arene 11a
A solution of carbene complex 1c (R1, R2 = Me, R3 = H) [21,22]
(0.1284 g, 0.465 mmol), alkyne 2 (0.075 g, 0.699 mmol), 2,6-luti-
dine (0.14 mL, 1.16 mmol), and TBSOTf (0.16 mL, 0.699 mmol) in
9.5 mL of CH2Cl2 was added to a 50 mL flame dried pear-shaped
single-necked flask in which the 14/20 joint was replaced by a high
vacuum T-shaped Teflon value. The system was deoxygenated by
the freeze–thaw method and after the third cycle, the flask was
back-filled with argon at room temperature. The flask was sealed
by closing the Teflon value and the flask was then heated at
50 °C under argon for 19 h. After removal of the solvents, the crude
product was purified by chromatography on silica gel with 1:1:4
mixture of ether/methylene chloride/hexane to give arene complex
11 (0.1596 g, 0.340 mmol, 73%) as a yellow solid. Spectral data for
11: (mp 96–99 °C, Rf = 0.21, 1:1:4) 1H NMR (CDCl3) d 0.31 (s, 3 H),
0.37 (s, 3 H), 0.97 (s, 9 H), 1.67–1.78 (m, 4 H), 2.14 (s, 3 H), 2.20–
2.24 (m, 1 H), 2.39 (t, 2 H, J = 6.0 Hz), 2.70–2.74 (m, 1 H), 3.69 (s,
3 H), 5.13 (s, 1 H), 5.25 (s, 1 H); 13C NMR (CDCl3) d ꢀ4.5, ꢀ4.4,
15.8, 17.0, 18.0, 25.1, 25.3, 29.6, 29.7, 56.5, 79.3 (C-2, Car-H), 86.3
(C-5), 97.8 (C-6), 100.3 (C-3), 119.3 (CN), 131.4 (C-4), 134.4 (C-1,
C-OMe), 234.7; IR (neat film) 2248w, 1952s, 1868s, 1484m,
1371m, 1224m, 1019m cmꢀ1; mass spectrum m/z (relative inten-
sity) 469 (M+, 5), 400 (15), 385(45), 370 (8), 333 (15), 276 (100),
248 (15). Anal. Calc. for C22H31CrNO5Si: C, 56.27; H, 6.65; N, 2.98.
Found: C, 56.14; H, 6.33; N, 2.91%.
C
19H31NO2Si m/z 333.2124, meas 333.2134.
The 13C, 2H-labeled 11a was obtained by stirring 13C, 2H-labeled
11 in CHCl3 in air for 24 h. Filtration through 1:1 silica gel/Celite
afforded a mixture of the desired free arene and the starting
arene–Cr(CO)3 complex. The aromatic carbons of the free arene
could be assigned from the spectrum of this mixture based on
the following evidence: d 146.6 (C-4, C-OTBS, increased signal
intensity), 151.9 (C-1, C-OMe, lack of increased signal intensity),
129.0 (C-3, C-H from unlabeled spectrum, surrounded by 13C–13
C
doublet, J ꢁ 70 Hz), 112.1 (C-2, Car-H, decrease in signal intensity
in 2H-labeled compound), 124.9 (C-6, lack of 13C–13C doublets),
121.0 (C-5, surrounded by 13C–13C doublet, J ꢁ 70 Hz).
2.6. Synthesis of arene chromium tricarbonyl complex 12 and the free
arene 12a
A solution of carbene complex 1d (R1 = c-C6H11, R2 = Me, R3 = H)
[23] (0.5319 g, 1.545 mmol), alkyne 2 (0.248 g, 2.32 mmol), 2,6-lu-
tidine (0.45 mL, 3.9 mmol), and TBSOTf (0.53 mL, 2.3 mmol) in
30 mL of CH2Cl2 was added to a 100 mL flame dried pear-shaped
single-necked flask in which the 14/20 joint was replaced by a high
2.5.1. Assignment of the arene carbons of complex 11
The 13C, 2H-labeled arene–Cr(CO)3 complex 11 was prepared as
follows: A solution of carbene complex 1c (0.27 g, 1.0 mmol) in
5 mL CH2Cl2 was degassed by the freeze–pump–thaw method
(three cycles). After the final cycle the reaction flask was charged
with 1 atm 99% 13CO, sealed, and heated for 18 h at 50 °C. The flask
was then charged with alkyne 2 (0.075 g, 0.70 mmol) which had
Table 4
Variable temperature 13C NMR data for complex 11 (CD2Cl2).
Temp (°C)
CO
C-1
C-4
C-3
C-6
C-5
C-2
20
0
ꢀ25
ꢀ50
ꢀ60
234.80
234.80
234.76
234.71
234.68
134.45
134.32
134.13
133.93
133.83
131.45
131.22
130.90
130.55
130.42
100.66
100.57
100.46
100.33
100.28
97.95
97.77
97.53
97.26
97.16
86.49
86.44
86.34
86.25
86.29
79.59
79.39
79.15
78.92
78.85
the alkenyl proton replaced with
a deuteron, 2,6-lutidine
(0.29 mL, 2.5 mmol), and TBS-triflate (0.35 mL, 1.5 mmol). After
degassing the reaction flask was again charged with 1 atm of

212
S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
vacuum T-shaped Teflon value. The system was deoxygenated by
the freeze–thaw method and after the third cycle, the flask was
back-filled with argon at room temperature. The flask was sealed
by closing the Teflon value and the flask was then heated at
50 °C under argon for 23 h. After removal of the solvents, the crude
product was purified by chromatography on silica gel with 1:1:10
mixture of ether/methylene chloride/hexane followed by crystalli-
zation from ether/pentane to give arene complex 12 (0.6244 g,
1.161 mmol, 75%) as golden yellow prisms. Spectral data for 12:
(mp 87–89 °C, Rf = 0.46, 1:1:4) 1H NMR (CDCl3) d 0.32 (s, 3 H),
0.38 (s, 3 H), 0.98 (s, 9 H), 1.30–1.45 (m, 3 H), 1.51–1.61 (m, 3
H), 1.70–1.82 (m, 6 H), 1.88–1.94 (m, 1 H), 1.95–2.00 (m, 1 H),
2.15 (s, 3 H), 2.20–2.25 (m, 1 H), 2.40 (t, 2 H, J = 6.0 Hz), 2.68–
2.75 (m, 1 H), 3.98–4.03 (m, 1 H), 5.15 (s, 1 H), 5.19 (s, 1 H); 13C
NMR (CDCl3) d ꢀ4.3, ꢀ4.2, 16.1, 17.1, 18.1, 23.3, 23.4, 25.27,
25.32, 25.5, 29.8, 29.9, 31.6, 32.0, 78.1, 81.5, 86.5, 98.0, 100.6,
119.2, 131.1, 133.4, 234.9; IR (neat film) 2250vw, 1948vs,
1869vs, 1480m, 1371m, 1222m, 841m, 674m cmꢀ1; mass spec-
trum m/z (rel. intensity) 537 (M+, 10), 453(95), 401 (15), 370
(55), 344 (5), 313 (5), 262 (100), 231 (10). Anal. Calc. for
NMR (CDCl3) d –2.9, –2.6, 17.1, 17.8, 18.6, 25.1, 25.8, 28.8, 29.9,
55.7, 78.0 (C-2/6), 78.8 (C-2/6), 102.1(C-3/5), 106.0 (C-3/5), 119.2
(CN), 126.4 (C-4), 139.2 (C-1, C-OMe), 234.4; IR (neat film)
2247w, 1955s, 1862s, 1546m, 1465s, 1264m, 1050m cmꢀ1; mass
spectrum m/z (rel. intensity) 469 (M+, 10), 385 (70), 370 (10),
328 (20), 276 (40). Anal. Calc. for C22H31CrNO5Si: C, 56.27; H,
6.65; N, 2.98. Found: C, 56.35; H, 6.82; N, 2.92%. The aromatic car-
bon signals were assigned on the following evidence. C-1 and C-4
by analogy to the three 13C-labeled complexes 11, 9, and 10 in
which the C-1 resonance is always higher field than the C-4 reso-
nance and C2/6 by the large signal intensity indicating Car–H.
2.7.1. Synthesis of the free arene 13a
Iodine (0.43 g, 1.7 mmol, 5 equiv) was added as a solid to a stir-
red solution of arene–Cr(CO)3 complex 13 (0.160 g, 0.341 mmol) in
2 mL THF under air. TLC indicated 13 had been consumed in less
than 10 min with clean conversion to 13a. The mixture was diluted
with ether, washed with 10% Na2S2O3, 1% HCl, and brine, then
dried over MgSO4. Filtration through a short plug of 1:1 silica
gel/Celite and removal of solvent afforded arene 13a (yield not
determined) which was characterized without further purification.
Spectral data for 13a: (pale yellow oil, Rf = 0.48, 1:1:4) 1H NMR
(CDCl3) d 0.18 (s, 6 H), 1.03 (s, 9 H), 1.62–1.74 (m, 4 H), 2.20 (s, 3
H), 2.32 (t, 2 H, J = 7.0 Hz), 2.60 (t, 2 H, J = 7.2 Hz), 3.74 (s, 3 H),
6.48 (d, 1 H, J = 3.0 Hz), 6.52 (d, 1 H, J = 3.1 Hz); 13C NMR (CDCl3)
d ꢀ3.2, 16.9, 18.1, 18.6, 24.8, 26.0, 29.0, 29.9, 55.3, 112.6 (C-2 or
6), 113.9 (C-2 or 6), 119.6, 129.4, 132.4, 145.4, 153.5; IR (neat film)
2246w, 1606w, 1482sh, 1472s, 1441sh, 1254sh, 1219s,
1066m cmꢀ1; mass spectrum m/z (rel. intensity) 333 (M+, 60),
318 (15), 276 (100), 257 (5), 248 (50), 234 (5), 219 (85), 207
(20), 195 (25), 183 (85), 175 (20), 163 (10), 151 (85); HRMS calc
for C19H31NO2Si m/z 333.2124, meas 333.2122.
C27H39CrNO5Si: C, 60.31; H, 7.31; N, 2.60. Found: C, 59.94; H,
7.14; N, 2.38%.
2.6.1. Synthesis of the free arene 12a
Iodine (0.32 g. 1.26 mmol, 4 equiv) was added as a solid to a
solution of arene–Cr(CO)3 complex 12 (0.166 g, 0.314 mmol) dis-
solved in 2 mL THF stirred at rt under air. After 10 min the reaction
mixture was diluted with 2 mL CHCl3. After 5 min TLC showed con-
sumption of 12 with the formation of several additional com-
pounds. The mixture was washed with 10% Na2S2O3, brine, and
dried with MgSO4. Concentration and chromatography (1:1:10)
afforded arene 12a (0.032 g, 0.080 mmol, 26%). Spectral data for
12a: (colorless oil, Rf = 0.64, 1:1:4) 1H NMR (CDCl3) d 0.23 (s, 6
H), 1.02 (s, 9 H), 1.32–1.41 (m, 3 H), 1.52–1.60 (m, 3 H), 1.65–
1.82 (m, 6 H), 1.92–1.97 (m, 2 H), 2.17 (s, 3 H), 2.34 (t, 2 H,
J = 6.7 Hz), 2.57 (t, 2 H, J = 6.9 Hz), 4.04–4.09 (m, 1 H), 6.55 (s, 1
H), 6.59 (s, 1 H); 13C NMR (CDCl3) d ꢀ4.2, 16.2, 17.1, 18.2, 23.7,
25.0, 25.7, 25.8, 29.2, 29.6, 32.1, 76.7, 116.7, 119.7, 120.8, 126.8,
129.0, 146.9, 149.9; IR (neat film) 2246w, 1501m, 1402m,
1254w, 1208m, 1048w, 967m cmꢀ1; mass spectrum m/z (rel.
intensity) 401 (M+, 85), 386 (10), 371 (5), 344 (45), 333 (15), 318
(10), 276 (75), 262 (100), 248 (10), 234 (30), 219 (10), 207 (10),
193 (35), 181 (20), 165 (10), 147 (15); HRMS calc for C24H39NO2Si
m/z 401.2750, meas 401.2767.
2.8. Synthesis of arene chromium tricarbonyl complex 14 and the free
arene 14a
A solution of carbene complex 1e (R1 = Me, R2 = H, R3 = Me) [24]
(0.709 g, 2.57 mmol), alkyne 2 (0.413 g, 3.85 mmol), 2,6-lutidine
(0.75 mL, 6.42 mmol), and TIPSOTf (1.04 mL, 3.85 mmol) in 40 mL
of CH2Cl2 was added to a 100 mL flame dried pear-shaped single-
necked flask in which the 14/20 joint was replaced by a high vac-
uum T-shaped Teflon value. The system was deoxygenated by the
freeze–thaw method and after the third cycle, the flask was back-
filled with argon at room temperature. The flask was sealed by
closing the Teflon value and the flask was then heated at 60 °C un-
der argon for 25 h. After removal of the solvents, the crude product
was purified by chromatography on silica gel with 1:1:4 mixture of
ether/methylene chloride/hexane followed by crystallization from
ether to give arene–Cr(CO)3 complex 14 (0.899 g, 1.76 mmol,
69%) as yellow prisms. Spectral data for 14: (mp 95–97 °C,
Rf = 0.40, 1:1:4) 1H NMR (CDCl3) d 1.14 (d, 18 H, J = 7.0 Hz), 1.21–
1.30 (m, 3 H), 1.80–1.91 (m, 4 H), 2.30 (s, 3 H), 2.41–2.51 (m, 2
H), 2.70 (t, 2 H, J = 6.0 Hz), 3.64 (s, 3 H), 5.02 (d, 1 H, J = 2.5 Hz),
5.10 (d, 1 H, J = 2.5 Hz); 13C NMR (CDCl3) d 14.24, 17.10, 17.93,
18.40, 25.54, 27.42, 28.88, 55.89, 78.62 (C-2), 79.36 (C-6), 99.06
(C-5), 104.36 (C-3), 119.33 (CN), 131.38 (C-4), 137.93 (C-1),
234.41; IR (neat film) 2248vw, 1952s, 1869sh, 1866s, 1546w,
1465m, 1424m, 1235m, 1050w, 882m cmꢀ1; mass spectrum m/z
(rel. intensity) 511 (M+, 5), 427 (60), 375 (20), 332 (100), 300
(10), 270 (5). The carbon at C-4 was assigned based on enhanced
intensity in the 13C labeled compound and C-3 and C-5 were iden-
tified from their coupling to C-4. The rest of the carbons were as-
signed based on a gHMBC experiment in which the methyl
singlet at 2.30 was found to be coupled to C-4, C-5 and C-6, but
not to C-1, C-2 or C-3. The 13C NMR spectrum of complex 14 was
also investigated at a number of temperatures down to ꢀ60 °C in
2.7. Synthesis of arene chromium tricarbonyl complex 13 and the free
arene 13a
A solution of carbene complex 1e (R1 = Me, R2 = H, R3 = Me) [24]
(0.6419 g, 2.326 mmol), alkyne 2 (0.374 g, 3.49 mmol), 2,6-lutidine
(0.68 mL, 5.82 mmol), and TBSOTf (0.80 mL, 3.49 mmol) in 25 mL
of CH2Cl2 was added to a 100 mL flame dried pear-shaped single-
necked flask in which the 14/20 joint was replaced by a high vac-
uum T-shaped Teflon value. The system was deoxygenated by the
freeze–thaw method and after the third cycle, the flask was back-
filled with argon at room temperature. The flask was sealed by
closing the Teflon value and the flask was then heated at 65 °C un-
der argon for 20 h. After removal of the solvents, the crude product
was purified by chromatography on silica gel with 1:1:4 mixture of
ether/methylene chloride/hexane followed by crystallization from
ether/pentane to give the arene–Cr(CO)3 complex 13 (0.7982 g,
1.70 mmol, 73%). Spectral data for 13: (fine yellow needles
[ether/pentane], mp 77–78 °C, Rf = 0.29, 1:1:4) 1H NMR (CDCl3) d
0.357 (s, 3 H), 0.362 (s, 3 H), 1.01 (s, 9 H), 1.73–1.89 (m, 4 H),
2.24 (s, 3 H), 2.31–2.39 (m, 1 H), 2.42 (t, 2 H, J = 5.9 Hz), 2.77–
2.83 (m, 1 H), 3.67 (s, 3 H), 4.943 (s, 1 H), 4.945 (s, 1 H); 13C

S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
213
Table 5
We chose to look at the differences in shifts between the com-
plexed and uncomplexed arenes to compare arenes with different
substitution patterns. Comparison of absolute chemical shift values
of differently substituted arene–Cr(CO)3 complexes could be com-
plicated by substituent effects. Bodner and Todd [25] have demon-
strated that resonance effects of ring substituents are nearly
unchanged by –Cr(CO)3 complexation, while inductive effects are
only slightly altered, so differences between the spectra of the
complexed and free arenes can be attributed as primarily resulting
from the metal fragment. Since the oxygen substituents are para
any inductive or resonance effects on other ring carbons should
be offsetting. It is important to note that the meta carbons of tolu-
ene-Cr(CO)3 are upfield shifted more than the methyl-substituted
carbon (6.3 ppm, 19%), though the ortho and para carbons are
shifted even farther upfield. This means that differences in the
amount of upfield shift between similarly substituted 3,6 and 5,2
carbons can be assigned more reliably as the result of the preferred
conformation of the metal fragment than it can be for dissimilarly
substituted (methyl versus proton) pairs of ring carbons. Because
of this, DDd values obtained from dissimilarly substituted carbons
are either placed in brackets or ‘‘corrected” by adjusting the C–ar–
H resonance by 6 ppm.
Variable temperature 13C NMR data for complex 14 (CD2Cl2).
Temp (°C)
CO
C-1
C-4
C-3
C-5
C-6
C-2
20
0
ꢀ25
ꢀ50
ꢀ60
–
–
–
–
–
–
–
234.43
234.35
234.26
234.22
137.88
137.80
137.71
137.70
131.07
130.73
130.29
130.25
104.95
105.06
105.19
105.26
99.56
99.64
99.74
99.78
78.51
78.67
78.18
77.99
78.48
78.22
77.91
77.99
CD2Cl2 and the data are presented in Table 5. Six signals were iden-
tified in the aromatic region of the solid-state 13C NMR spectrum
that correlated well with the chemical shifts in solution. The
assignments made for these carbons are: C-1 (139.8), C-4 (136.6),
C-3 (108.5), C-5 (100.2), C-6 (81.1), C-2 (77.0).
2.8.1. Synthesis of the free arene 14a
Iodine (0.32 g, 1.28 mmol, 4 equiv) was added as a solid to a
solution of arene–Cr(CO)3 complex 14 in 3 mL CHCl3 stirred in
air. TLC showed a fairly clean conversion to 14a. After 15 min the
mixture was washed with 10% Na2S2O3, then dried with brine
and MgSO4. Concentration and chromatography (1:1:10) afforded
arene 14a (yield not determined) as a colorless solid. Spectral data
for 14a: (colorless solid, mp 69–71 °C, Rf = 0.51, 1:1:4) 1H NMR
(CDCl3) d 1.12 (d, 18 H, J = 7.4 Hz), 1.26–1.33 (m, 3 H), 1.68–1.78
(m, 4 H), 2.25 (s, 3 H), 2.34 (t, 2 H, J = 6.8 Hz), 2.63 (t, 2 H,
J = 7.2 Hz), 3.74 (s, 3 H), 6.47 (d, 1 H, J = 2.8 Hz), 6.50 (d, 1 H,
J = 2.9 Hz); 13C NMR (CDCl3) d 14.26, 17.08, 17.99, 18.27, 25.17,
29.27, 30.26, 55.36, 112.41 (C-2), 113.83 (C-6), 119.60 (CN),
128.93 (C-5), 132.00 (C-3), 146.82 (C-4), 153.35 (C-1); IR (neat
By comparing the differences that result from complexation in
the relative positions of carbons para to one another, it should be
possible to see if the 1,3,5 or 2,4,6 carbons are shifted farther up-
field and deduce the average preferred conformation of the metal
fragment. The difference in absolute chemical shift (Dd) for car-
bons para to one another is used, since for the complexes in this
study carbons para to one another tend to be symmetrically substi-
tuted with ether, alkyl, or proton substituents, so DDd values
should be only minimally effected by whatever substituent effects
there might be. An equation for obtaining DDd values is shown in
Eq. (1) [13c]. Cn represents C-1,3 or 5, while Cn+3 represents a cor-
responding para carbon, C-4, 6 or 2, respectively.
film) 2245w, 1594w, 1480m, 1466m, 1220s, 1069w, 994m cmꢀ1
;
mass spectrum m/z (rel. intensity) 375 (M+, 60), 332 (M–i-Pr,
100), 304 (20), 290 (10), 276 (5), 262 (5), 246 (5), 234 (5), 219
(8), 207 (10), 193 (10), 179 (5), 166 (15); HRMS calc for C22H37NO2-
Si m/z 375.2594, meas 375.2566.
The carbon at C-4 was assigned based on enhanced intensity in
the 13C labeled compound and C-3 and C-5 were identified from
their coupling to C-4. The rest of the carbons were assigned based
on a gHMBC experiment in which the methyl singlet at 2.25 was
found to be coupled to C-4, C-5 and C-6 with only weak coupling
to C-3 and C-2. This was confirmed by gHMBC in that the triplet
at 2.63 was coupled to the C-3, C-4 and C-2 but not C-6, C-5 or
C-1.
DDd ¼
D
d½Arene—CrðCOÞ3ðdCn ꢀ dCnþ3Þꢃ ꢀ
D
d½AreneðdCn ꢀ dCnþ3Þꢃ
ð1Þ
An example of this type of analysis for arene–Cr(CO)3 complex 9
is shown in Table 6. Values for
Dd were obtained from the assigned
resonances of the 13C NMR spectra obtained at ambient tempera-
ture in CDCl3 for both arene–Cr(CO)3 complex 9 and the corre-
sponding free arene 9a. As can be seen the
Dd values for the
pairs of para carbons are generally larger in the –Cr(CO)3 complex
than they are in the uncomplexed arene. This means that, for
example, C-4 is shifted upfield (shielded) in the arene–Cr(CO)3
complex by more than C-1 is shifted upfield. This can be inter-
preted as evidence that, on average, C-4 is eclipsed less often by
a CO ligand than C-1. This in turn would imply that C-2 would
be eclipsed less often than C-3, leaving C-3 more activated for
nucleophilic attack than C-2. The DDd values for C-5/C-2 is a posi-
tive number, while C-3/C-6 is slightly negative. The average of the
three DDd values is +5.3. This implies that on average there is a
preferred eclipsing of the 1,3,5 ring carbons. Even if the value
DDd C-5/C-2 is adjusted by subtracting the ca. 6 ppm upfield shift
to be expected for the unsubstituted C-2, a value of ca. +3.3 is ob-
tained. When the nucleophilic addition reactions are carried out at
ꢀ78 °C, the only products of cyclization we have isolated are those
where attack has occurred at C-3.
As shown in Table 7, similar analysis of the three arene–Cr(CO)3
complexes which differ from 9 only in the pattern of alkyl substit-
uents on the arene ring shows a correlation between the sign of the
DDd value and the products obtained after ꢀ78 °C nucleophilic
addition. Complexes 13 and 8 have DDd values which would cor-
respond to the configuration of the metal fragment deduced from
analysis of the conformers in Scheme 3. Again even after adjusting
3. Results and discussion
3.1. Correlation of carbon-13NMR and position of kinetic attack
NMR spectra of kinetically-restricted rotational systems can
show a set of resonances for each conformer present. Thermody-
namically-restricted rotational systems show a single set of reso-
nances which is a weighted average of the conformer population.
The effect of –Cr(CO)3 complexation to an arene is a large (20–
35 ppm) upfield shift in the aromatic resonances. In systems where
there is a preferred –Cr(CO)3 conformation, those carbons that are
staggered are shifted upfield (shielded) more than those that are
eclipsed. This is consistent with the analysis of Albright [12a] in
which the arene HOMO coefficients are larger at ring carbons stag-
gered by the Cr(CO)3 fragment. That the arene–Cr(CO)3 complexes
in this study are thermodynamically-restricted rotational systems
and, hence, that the –Cr(CO)3 fragment is activating either the ipso
or vicinal position can be deduced from the 13C NMR spectra of
these complexes. This is done by comparing the difference (DDd)
in magnitude of the upfield shift (Dd) in the various aromatic res-
onances upon complexation of the free arene to the metal
fragment.

214
S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
Table 6
haves similarly, but that the relatively smaller C-3 alkyl tether in
avoiding this large group is forced to interact with the metal frag-
ment. That alkyl substituents forced syn to the metal fragment can
effect its preferred configuration has been demonstrated in several
systems [13,16,26].
Analysis of DDd values of arene-Cr(CO)3 complex 9/arene 9a.
The DDd values for complexes 10, 14, and 12 (Table 8) show a
strong correlation for complex 10, a possible correlation with 14,
while values for the cyclohexyl complex 12 indicate no clearly pre-
ferred configuration for the metal fragment. It is interesting to note
that the cyclohexyl complex gave the least efficient nucleophilic
addition among those studied in this work. The contrast between
the two TIPS-protected complexes helps to explain the difference
in the observed position of kinetic attack. It is possible that for
10 (DDd < 0) the large TIPS-group, certain to be above the plane
of the arene and above the face opposite the metal fragment,
causes the alkyl tether to spend relatively more time on the same
side as the –Cr(CO)3 fragment, interfering with the CO ligands, and
leading to a preference of the C-2 eclipsed conformer. This would
explain the difference in reactivity of 10 as compared with the
TBS-protected 9.
13C NMR d (ppm)
Arene Cr(CO)3 complex
C-1
C-4
137.3
125.6
+11.7
C-3
C-6
102.4
98.9
+3.5
C-5
C-2
106.1
73.4
+32.7
D
d
D
d
D
d
Uncomplexed arene
C-1
C-4
151.7
144.8
+6.9
C-3
C-6
129.9
124.7
+5.2
C-5
C-2
Dd
127.7
107.9
+19.8
D
d
D
d
DDd = [Arene–Cr(Co)3Dd[C1,3,5–C4,6,2]] ꢀ [Arene
Dd[C1,3,5–C4,6,2]].
In contrast, the other TIPS-protected complex 14 (DDd > 0) has
no alkyl substituent at C-6. Despite this proposed effect on the C-3
alkyl tether, the C-3 eclipsed conformation could remain favored,
since all three eclipsed carbons have electron-releasing substitu-
ents compared to the C-2 eclipsed conformation which has only
one such stabilizing interaction. That the TIPS-oxy group itself does
not completely control the preferred conformation is analogous to
the fact that neo-pentyl benzene adopts a preferred configuration
of the metal fragment eclipsing the substituted carbon [25]. Of
course the actual reasons for the –Cr(CO)3 fragment adopting the
preferred configuration it does in each complex are not estab-
lished. Nonetheless the correlation between the 13C NMR data
and the observed position of kinetic attack, pointing to selective
activation of either the C-2 or C-3 position by the metal fragment,
seems clear.
The population of conformers can be approximated through Eq.
(2) [27]. In this equation PA stands for the percentage of the major
conformer, hmax represents the maximum change in chemical shift
between an eclipsed and staggered carbon, and h represents the
DDd value of the compound. This approximation assumes the pres-
ence in solution of only the two eclipsed conformers, the staggered
pi-bond conformer being seen as a relatively short-lived interme-
diate. This also assumes a linear relationship between the confor-
mational population and the weighted average in chemical shift.
C-1/C-4
C-3/C-6
C-5/C-2
+4.8
–1.7
[+12.9]a
a Uncorrected for difference in
Dd due to asymmetric H vs. alkyl substitution.
the DDd values of dissimilarly substituted carbon pairs by ca.
6 ppm, these values remain greater than zero. Since, the labeled
carbene complex necessary for the synthesis of 13C-labeled 8 could
not be prepared, the values indicated in Table 7 are based on ten-
tative assignments of C-2 and C-5 (see Section 2).
Especially striking is the negative DDd value for 11, the only
complex in this series which affords fused-ring products at
ꢀ78 °C. As mentioned above we could see no a priori reason to pre-
dict a preferred eclipsing of C-2. Perhaps with the electronic effects
of the ring substituents nearly equal, an adverse steric reaction be-
tween the C-3 alkyl chain and the –Cr(CO)3 fragment leads to the
preferred configuration indicated by these data. It has been shown
that the Cr(CO)3 fragment in neo-pentylbenzene preferentially
eclipses the substituted ring position in solution similarly to tolu-
ene-Cr(CO)3 which contrasts tert-butylbenzene-Cr(CO)3 [26]. The
neo-pentyl group is able to avoid interacting with the metal frag-
ment by spending most of its time above the face of the arene
opposite the metal. In 11 it is possible that the TBS-oxy-group be-
Table 7
Correlation between DDd values and position of kinetic attack.
Complex
R2
R3
DD
d
Position of kinetic attack
(C1-C4)
(C3-C6)
(C5-C2)
9
–(CH2)4–
Me
H
+4.8
ꢀ2.3
+4.7
+1.2
ꢀ1.7
[+12.9]b
ꢀ2.0
C-3
C-2
C-3
C-3
11
13
8
H
Me
H
ꢀ1.6
[+8.7]b
[+7.4]a,b
[+7.3]b
+2.1a
H
a
Based on tentative assignment of C-2 and C-5.
b
Uncorrected for difference in
Dd due to asymmetric H vs. alkyl substitution.

S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
215
Table 8
Correlation between DDd values and position of kinetic nucleophilic attack.
Complex
R1
R2
R3
R4
DDd (ppm)
Observed kinetic attack
(C1–C4)
(C3–C6)
(C5–C2)
14
10
12
Me
Me
Cyclo-hexyl
H
Me
H
TIPS
TIPS
TBS
+0.02
ꢀ3.5
ꢀ0.7
[+6.8]a
ꢀ2.3
[+4.0]a
[+4.4]a
+0.9
C-3
C-2
C-2
–(CH2)4–
Me
+0.4
a
Uncorrected for difference in
Dd due to asymmetric H vs. alkyl substitution.
A value of 9 ppm for alkyl substituted carbons can be assigned to
hmax from the work of McGlinchy and co-workers [18a]. The DD
values for the ambient temperature data for complexes 11, 13, 9,
10, and perhaps 14 can be approximated as falling in a range be-
tween 2 and 4 ppm. This would translate to approximate popula-
tions of conformers of between ca. 60:40 and 75:25.
Cr(CO)3 complexes over a temperature range from 23 to ꢀ60 °C
in CD2Cl2 at 75 MHz. The differences in the 13C resonances ob-
served are listed in Table 9. The aryl resonances of these complexes
at 20, 0, ꢀ25, ꢀ50, and ꢀ60 °C are included in the Section 2. In
addition the uncomplexed arene 9a was also examined over a sim-
ilar temperature range as a control.
d
The data in Table 9 are consistent with the analysis presented
above for three of the five complexes examined. Complexes 9, 8,
and 14 all are attacked kinetically at C-3. The resonances for C-1,
C-3, and C-5, upon cooling, are shifted downfield relative to the
C-4, C-6, and C-2 resonances, respectively. Since staggered carbons
are shielded relative to eclipsed carbons, these changes indicate an
increase in the population of the C-1,3,5 eclipsed conformation at
lower temperature. The amount of change in these resonances is
generally greater than that seen for 9a upon cooling, suggesting
the effect is the result of the presence of the metal fragment. The
data for complexes 11 and 10 show very little change in the aryl
carbon resonances on cooling, suggesting little change in the pop-
ulation of conformers. These two complexes are both attacked
kinetically at the C-2 position, though the reasons for the differ-
ence between these two complexes and those which are attacked
at C-3 are not clear.
PA ¼ 50 þ 50ðh=hmax
Þ
ð2Þ
3.2. Low temperature carbon-13NMR study of representative
complexes
In complexes with extremely sterically congested arene ligands
at low temperature, a splitting of the single 13C resonance of the
three CO ligands can become resolved due to rotation of the metal
sufficiently hindered for the CO carbons to become magnetically
inequivalent. It might also be anticipated that we might see
changes in the DDd values of the 1,3,5 versus 2,4,6 carbons if, upon
cooling, the –Cr(CO)3 fragment rotated through disfavored confor-
mations less frequently. We investigated several of the arene–
Table 9
The low temperature data from Table 9 can be combined with
the DDd values from Tables 7 and 8 to give approximate popula-
tions of conformers at –60 °C. These approximate values (Table 10)
are based on a simple average of the three DDd values for each
compound and are obtained directly from Eq. (2). The DDd values
for unsymmetrically substituted pairs of carbons have been cor-
rected by subtracting 6 ppm to correct for the difference in upfield
shift of proton versus alkyl substituted carbons. The slight
(0.1 ppm) difference in DDd values related to use of CD2Cl2 versus
CDCl3 has not been corrected for. Despite the fact that for 11 and
10 the low temperature spectra showed a shifting away from the
C-2,4,6 eclipsed conformation, the data show that in both com-
plexes this remains the more populated conformer. This is consis-
tent with the observed position of nucleophilic attack under
conditions favoring formation of kinetic addition products. The
reason for the weighted average of the ring carbons for these
two complexes to reflect an increase in the population of the C-
1,3,5 conformation similar to the other complexes examined is
not known.
Changes in chemical shift at low temperature for arene–Cr(CO)3 complexes 8–11 and
14 (CD2Cl2).
Compound
Temp (°C)
Dd
C1–C4
C3–C6
C5–C2
9
9
9a
9a
8
20
ꢀ60
20
11.8
13.9
7.0
7.2
7.6
8.6
6.8
7.5
3.0
3.4
1.9
2.0
4.2
5.5
5.3
32.6
33.9
–
ꢀ50
5.9
–
20
28.7
30.3
26.4
27.3
2.7
3.1
2.3
3.3
ꢀ1.2
ꢀ1.1
21.1
22.0
6.9
7.4
24.1
24.8
8
ꢀ60
20
14
14
11
11
10
10
ꢀ60
20
3.3. Correlation between proton NMR and position of kinetic attack
ꢀ60
23
ꢀ60
The hypothesis which correlates the 13C NMR data, position of
kinetic nucleophilic attack, and the Cr(CO)3 conformation is

216
S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
Table 10
3.4. Preferred conformations of the complexes 9, 14, 10, and 11 in the
solid-state
Approximate population of conformers in solution of arene–Cr(CO)3 complexes 8–11
and 14 (CD2Cl2).
There have been many instances where the position of intermo-
lecular nucleophilic attack has been correlated to the position of
the –Cr(CO)3 fragment in the solid-state structures of the arene–
Cr(CO)3 complexes. A danger associated with this type of analysis
is the fact that the structure observed in the solid-state does not
necessarily reflect the configuration of the same complex in solu-
tion. The Cr(CO)3 fragment in arene–Cr(CO)3 complexes is suscep-
tible to being twisted from its preferred solution conformation by
intermolecular packing forces. For example, in a study involving
the determination of a number of crystal structures of arene–
Cr(CO)3 complexes with donor substituents, it was reported that
the dihedral angles in anisole-Cr(CO)3 and 1,4-dimethoxyben-
zene-Cr(CO)3 were 8.8° and 18.8°, respectively [9i]. For this reason
we were reluctant to obtain solid-state structures of any of these
complexes, since information obtained regarding metal fragment
geometry can be open to question.
A report from the McGlinchy laboratories showed a close corre-
lation between the 13C solution and solid-state spectra of pentae-
thylacetophenone-Cr(CO)3 [18a]. We wondered if we would be
able to see a similar correlation and be able to define a more secure
relationship between the preferred conformation of the Cr(CO)3
fragment in solution and in the solid-state. Single crystals of com-
plexes 9, 14, 10, and 11 were grown and solid-state structures
were obtained by X-ray diffraction spectroscopy. In addition so-
lid-state 13C NMR spectra of ca. 120–150 mg each of single crystals
of 9, 14, and 11 were obtained. The solid-state 13C spectrum of 10
of sufficient quality could not be obtained.
Compound Temp
DDda
Avg. % C1,3,5
eclipsed
(°C)
C1–
C4
C3–
C6
C5–
C2
9
9
8
20
ꢀ60
20
+4.9
+7.0
+ 1.2
+2.2
+0.3
+1.0
ꢀ2.3
ꢀ1.9
ꢀ3.4
ꢀ3.3
+1.1
ꢀ0.4
+1.6b
+3.2b
+2.3b
+3.2b
ꢀ1.4
ꢀ1.0
ꢀ2.4
ꢀ1.4
+6.7b
+8.0b
+2.0
+3.4 69
+4.9 78
+1.6 59
+2.5 64
+0.4 52
+1.2 57
ꢀ1.9 39
ꢀ1.5 42
ꢀ2.4 37
ꢀ1.8 40
8
ꢀ60
+2.1
14
14
11
11
10
10
20
ꢀ1.5b
ꢀ0.6b
ꢀ2.0
ꢀ1.5
ꢀ1.4b
ꢀ0.7b
ꢀ60
20
ꢀ60
23
ꢀ60
a
Compared to 23 °C spectra of the corresponding free arene in CDCl3.
Corrected by subtracting 6 ppm for the additional shift of
substitution.
b
H vs. alkyl
supported by additional evidence found in a similar analysis of 1H
NMR data. The correlation between the difference in upfield shift
ORTEP diagrams of complexes 9, 14, 10 and 11 and the relevant
dihedral angles are presented in Figs. 1 and 2. For complex 9, 13C
NMR analysis in solution (CD2Cl2) correctly predicts that attack
should occur preferentially at C-3 with a DDd of +4.9 at –60 °C (Ta-
ble 10). The X-ray structure of complex 9 (Fig. 1) reveals that,
although one of the CO ligands is not directly eclipsing the C-3 car-
bon, it comes close with a dihedral angle of 9.7°. The closest CO li-
gand to the C-2 carbon of complex 9 has a dihedral angle of 34.7°.
Likewise 13C NMR analysis in solution (CD2Cl2) correctly predicts
that attack should occur preferentially in complex 14 at C-3 with
(DDd) of aryl protons, position of nucleophilic attack, and preferred
Cr(CO)3 conformation was first proposed by Rose to explain the
behavior of the spiroindane complex shown in Table 11 [13c]. As
is shown the staggered aryl protons (H-4, H-6) are shifted upfield
by an average of 24% more than the upfield shift of eclipsed aryl
protons (H-5, H-7). In both complexes 9 and 10 the aryl proton is
upfield-shifted slightly less than in the indane complex. This could
be because aryl resonances for the uncomplexed indane (d 7.05, s,
4H) begin so much farther downfield than the more electron-rich
uncomplexed hydroquinones 9a (d 6.41) and 10a (d 6.39). Support-
ing the idea that in 10 C-2 is preferentially eclipsed, while in 9 it is
preferentially staggered, the H-2 proton resonance for 9 is shifted
upfield by 17% more than the H-2 resonance of 10. Unlike the anal-
ysis of Rose, this analysis compares proton resonances from differ-
ent molecules, but the differences in upfield shift do seem to
approximate those seen in the spiroindane complex.
a
DDd of +1.2 at –60 °C (Table 10). The X-ray structure of complex
14 (Fig. 1) is consistent with this prediction since one of the CO li-
gands has a dihedral of 16.1° with C-3 while another has a dihedral
angle of 27.8° with C-2. In contrast to complexes 9 and 14, the 13
C
NMR data in solution predicts that attack in complex 10 with a
DDd of ꢀ1.8 at ꢀ60 °C (Table 10) should occur with C-2 and this
is what is observed (Scheme 2). This is supported by the crystal
structure of 10, albeit not strongly, since one of the CO ligands have
a smaller dihedral angle with C-2 (21.0°) than with C-3 (26.0°). The
most interesting structure is of complex 11 where the conforma-
tion of the metal fragment in the solid-state as determined by X-
ray diffraction is contrary to what was deduced in solution. The
13C NMR spectrum of complex 11 in solution correctly predicts that
attack should occur at C-2 since, as indicated in Table 10, DDd is
negative (ꢀ1.5 at ꢀ60 °C).
Table 11
1H NMR analysis of Cr(CO)3 conformation.
However, the X-ray structure of complex 11 (Fig. 2) shows that
in the solid-state that a CO ligand is much closer to C-3 (dihedral
angle 16.8°) than to the C-2 carbon (dihedral angle 31.2°).
The results from the 13C NMR spectra of complexes 9, 11 and 14
in both the solution and solid-state are presented in Table 12. Com-
plexes 9 and 14 (Fig. 1), while adopting somewhat staggered posi-
tions with respect to the C-2 and C-3 positions (numbering as
above C1-OMe, C3-tethered nitrile) are still relatively close to
eclipsing C3, the conformer deduced from their solution spectra
to be more heavily populated. We expected that the solid-state
13C NMR spectra of complexes 9 and 14 to be fairly similar to their
H-4
H-5
DDd = 0.27
D
D
d = 1.65
d = 1.38
H-6
H-7
DDd = 0.54
D
D
d = 1.82
d = 1.28
9 R4 = TBS,
10 R4 = TIPS,
DDd = 0.27 (17%)
D
d = 1.55
d = 1.28
D
Average max upfield shift: 1.74 ppm.
Average DDd = 0.41 (24%).

S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
217
arene-Cr(CO)3 complex 9
arene-Cr(CO)3 complex 14
C
23-Cr-C3-C12
C
C
23-Cr-C2-H
25-Cr-C10-C9
C7-Cr-C1-C10
C9-Cr-C5-O5
C8-Cr-C3-C15
Average
C7-Cr-C2-O4
C9-Cr-C6-H
C8-Cr-C4-H
Average
9.7°
9.0°
10.8°
9.8°
34.7°
33.4°
31.9°
33.3°
16.1°
10.5°
14.2°
13.6°
32.5°
27.8°
32.2°
30.8°
C25-Cr-C1-O1
C24-Cr-C5-C6
Average
C24-Cr-C4-O2
Average
Dihedral Angles (Crystallographic numbering)
Dihedral Angles (Crystallographic numbering)
Fig. 1. ORTEP diagrams of arene–Cr(CO)3 complexes 9 and 14.
arene-Cr(CO)3 complex 10
arene-Cr(CO)3 complex 11
C17-Cr-C5-H
16-Cr-C3-C7
C18-Cr-C1-O4
Average
C17-Cr-C6-C11
C16-Cr-C4-O5
C18-Cr-C2-C10
Average
C1-Cr-C5-C16
C3-Cr-C7-O5
C2-Cr-C9-H
Average
C1-Cr-C4-O4
C3-Cr-C6-H
C2-Cr-C8-C22
Average
21.0°
16.3°
22.2°
19.8°
26.0°
21.5°
25.3°
24.3°
16.8°
19.5°
16.3°
17.5°
27.0°
31.2°
31.4°
29.9°
C
Dihedral Angles (Crystallographic numbering)
Dihedral Angles (Crystallographic numbering)
Fig. 2. ORTEP diagrams of arene–Cr(CO)3 complexes 10 and 11.

218
S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
Table 12
that the eclipsed carbons of the predominant conformer are those
which suffer nucleophilic addition, it seems reasonable to propose
that in the intramolecular reactions the more favored conformer is
also the more reactive (lower energy transition state) since the car-
bons eclipsed by CO should be more electron depleted. The result-
ing DDGà would reflect the sum of the two energy differences
Comparison of solid-state and solution 13C NMR DD
complexes 11, 9 and 14.
d values for arene–Cr(CO)3
(DGrot and D
Gà) thereby telescoping the influence of the relative
energy differences of the two conformers with the result that the
differences of the energies of the transition states leading to the
two products could be sufficiently large (ꢁ2 kcal) to account for
the selective formation (>12:1) of the addition products observed
in this work under kinetic conditions.
Compound
Conditions
DDda
Avg.
C1–C4
C3–C6
C5–C2
9
9
9
14
14
14
11
11
11
20 °C, CD2Cl2
ꢀ60 °C, CD2Cl2
solid
20 °C, CD2Cl2
ꢀ60 °C, CD2Cl2
Solid
20 °C, CD2Cl2
ꢀ60 °C, CD2Cl2
Solid
+4.9
+7.0
+6.3
+0.3
+1.0
+1.7
ꢀ2.3
ꢀ1.9
+5.2
ꢀ1.1
ꢀ0.4
+1.0
+6.7b
+8.0b
+3.6b
ꢀ1.5b
ꢀ0.6b
+0.6b
ꢀ2.0
ꢀ1.5
+3.2
+3.4
+4.9
+3.6
+0.4
+1.2
+1.9
ꢀ1.9
ꢀ1.5
+5.9
4. Conclusions
+2.3b
+3.2b
+3.3b
ꢀ1.4
ꢀ1.0
+9.4
In a series of 1,4-dioxygenated arene chromium tricarbonyl
complexes, the position of kinetic attack by a nitrile-stabilized
carbanion tethered at C-3 by a three methylene tether was exam-
ined for four patterns of alkyl substitution at the 5 and 6 positions.
Generating and maintaining the carbanion at ꢀ78 °C for between 1
and 1.5 h, followed by addition of iodine at the same temperature
afforded good selectivity for generating addition products resulting
from addition at either C-2 or C-3. The position of kinetic attack
were correlated with ambient and low temperature 13C NMR spec-
tra, solid-state 13C NMR and crystallographic data, and 1H NMR
spectra. All of these data indicate an average preferred configura-
tion of the Cr(CO)3 fragment. It has been established that when
the Cr(CO)3 fragment adopts a preferred configuration, it is the pre-
dominate force in directing addition by nucleophiles. However, it is
the conformation of the Cr(CO)3 in solution that correlates with the
regioselectivity, not the conformation in the solid-state. We have
proposed that the high degree of regioselectivity in the intramolec-
ular addition reaction is the result of an additive electronic effect of
the alkyl substituents which both favors a preferred conformation
of the Cr(CO)3 fragment and lowers the energy of activation for
addition to the major conformer relative to the minor conformer.
a
Compared to 23 °C spectra of free arenes 9a, 14a, 11a in CDCl3.
Corrected by subtracting 6 ppm for the additional shift of
substitution.
b
H
vs. alkyl
low-temperature solution spectra and they are (Table 12). The X-
ray structure of complex 11 indicated that a different conformation
of the -Cr(CO)3 is preferred in the solid-state than in the solution
state as determined by 13C NMR analysis in CD2Cl2. Thus it was
anticipated that the solid-state spectrum of 11 would be signifi-
cantly different from that taken in solution with DDd values great-
er than zero. Indeed, this was found to be the case. The average
DDd value for the 13C NMR spectrum of complex 11 in solution
is ꢀ1.5, whereas, the average DDd for the 13C NMR spectrum of
complex 11 in the solid-state is +5.9 (Table 12). Thus, the example
of complex 11 indicates the shortcomings associated with predict-
ing the position of nucleophilic attack from the solid structure of
an arene chromium tricarbonyl complex.
Acknowledgment
This work was supported by a Grant from the National Science
Foundation (CHE-0750319).
3.5. Regioselectivity from additive electronic effects
The selectivity of kinetic nucleophilic attack seen with several
of the complexes discussed above was very good. The intramolec-
ular nucleophilic addition for complexes 11, 13, 9, 10, and 8 was
carried out at ꢀ78 °C so that the minor addition product was not
seen. Only with complex 14 was a fraction isolated which could
have contained a significant amount (12–15%) of a minor addition
product, which would have a selectivity of ca. 3:1. With the other
complexes selectivity was at least 12:1 (60%:5%) in favor of the
main product. This type of selectivity implies a difference in free
Appendix A. Supplementary material
CCDC 790425, 790426, 790427, and 790428 contain the supple-
mentary crystallographic data for this paper. These data can be ob-
tained free of charge from The Cambridge Crystallographic Data
Centre via http://www.ccdc.cam.ac.uk/data_request/cif. Supple-
mentary data associated with this article can be found, in the on-
line version, at doi:10.1016/j.ica.2010.07.037.
*
energy between the transition states (DDG ) leading to the major
References
and minor products of at least 1.5–2.0 kcal/mol. The conforma-
tional preferences indicated from the spectral data, primarily the
result of electronic effects do not seem sufficiently large to be
the only factor controlling the position of kinetic attack. The differ-
ence in the ground state energy of the two eclipsed conformers is
probably greater than 0.5 kcal/mol, but less than 1.0 kcal/mol and
with a rotational barrier roughly equivalent to that value [28].
Semmelhack and co-workers [10] have suggested that the tran-
sition states in nucleophilic addition of stabilized carbanions to
arene–Cr(CO)3 complexes are ‘‘late” or more product-like in nature.
If the relative stability of ‘‘product-like” transition states is consid-
ered, it can be seen that the effects would be additive, allowing for
[1] M.F. Semmelhack, H.T. Hall, M. Yoshifuji, G. Clark, J. Am. Chem. Soc. 97 (1975)
1247.
[2] K.H. Dötz, Angew. Chem., Int. Ed. 14 (1975) 644.
[3] E.O. Fischer, K. Ofele, Chem. Ber. 90 (1957) 2532.
[4] E.O. Fischer, A. Maasbol, Angew. Chem., Int. Ed. 3 (1964) 580.
[5] For recent reviews of this reaction, see: (a) M.R. Semmelhack, in: E.W. Abel,
F.G.A. Stone, G. Wilkinson (Eds.), Comprehensive Organometallic Chemistry
II, vol. 12, Pergamon, Oxford, UK, 1995, pp. 979–1016;
(b) F. Rose-Munch, E. Rose, in: D. Astruc (Ed.), Modern Arene Chemistry, Wiley-
VCH Verlag, 2002, pp. 368–399;
(c) M.F. Semmelhack, A. Chlenov, in: E.P. Kundig (Ed.), New Aspects of
Transition Metal Chemistry, Springer-Verlag, Berlin, 2004, p. 43;
(d) E.P. Kundig, A. Pape, Top. Organomet. Chem. 7 (2004) 157.
[6] For recent reviews of this reaction, see: (a) M.L. Waters, W.D. Wulff, Org. React.
70 (2008) 121;
*
a sufficiently large DDG to lead to the observed level of selectivity
[29]. Since it has been established in the intramolecular reaction
(b) K.H. Dötz, J. Stendel Jr., Chem. Rev. 109 (2009) 3227.

S. Chamberlin et al. / Inorganica Chimica Acta 364 (2010) 205–219
219
[7] S. Chamberlin, W.D. Wulff, J. Am. Chem. Soc. 114 (1992) 10667.
[8] S. Chamberlin, W.D. Wulff, B. Bax, Tetrahedron 49 (1993) 5531.
[9] (a) M.F. Semmelhack, G. Clark, J. Am. Chem. Soc. 99 (1977) 1675;
(b) M.F. Semmelhack, J.J. Harrison, Y. Thebtaranonth, J. Org. Chem. 44 (1979)
3275;
[15] (a) M.F. Semmelhack, L. Keller, Y. Thebtaranonth, J. Am. Chem. Soc. 99 (1977)
959;
(b) M.F. Semmelhack, J. Bisaha, M. Czarny, J. Am. Chem. Soc. 101 (1979) 768;
(c) M.F. Semmelhack, A. Yamashita, J. Am. Chem. Soc. 102 (1980) 5924;
(d) M.F. Semmelhack, P. Knochel, T. Singleton, Tetrahedron Lett. 34 (1993)
5051;
(e) L. Besson, M. Le Bail, D.J. Aitken, H.-P. Husson, F. Rose-Munch, E. Rose,
Tetrahedron Lett. 37 (1996) 3307.
(c) K. Schollkopf, J.J. Stezoski, F. Effenberger, Organometallics 4 (1985) 922;
(d) F. Camps, J. Coll, J.M. Moreto, L. Pages, J. Roget, J. Chem. Res. (Miniform)
(1990) 1824 (Synopses (1990) 236);
(e) E.P. Kundig, D. Amurrio, R. Liu, A. Ripa, Synlett (1991) 657;
(f) T.A. Albright, Tetrahedron 38 (1982) 1339;
[16] M. Uemura, H. Nishimura, T. Minami, Y. Hayashi, J. Am. Chem. Soc. 113 (1991)
5402.
(g) E.L. Muetterties, J.R. Bleeke, E.J. Wucherer, T.A. Albright, Chem. Rev. 82
(1982) 499;
(h) B. Ohlsson, C. Ullenius, J. Organomet. Chem. 350 (1988) 35;
(i) A.D. Hunter, L. Shilliday, W.S. Furey, M.J. Zawortko, Organometallics 11
(1992) 1550;
[17] (a) Y. Shvo, E.C. Taylor, K. Mislow, M. Raban, J. Am. Chem. Soc. 89 (1967) 4910;
(b) K. Mislow, M. Raban, in: N.L. Allinger, E.L. Eliel (Eds.), Topics in
Stereochemistry, vol. 1, John Wiley and Sons, New York, 1967, p. 1.
[18] (a) P.A. Downton, B. Mailvaganam, C.S. Frampton, B.G. Sayer, M.J. McGlinchy, J.
Am. Chem. Soc. 112 (1990) 27;
(j) C.H. Suresh, N. Koga, S.R. Cadre, Organometallics 19 (2000) 3008.
[10] M.F. Semmelhack, J.L. Garcia, D. Coates, R. Farina, R. Hong, B.K. Carpenter,
Organometallics 2 (1983) 467.
(b) K.V. Kilway, J.S. Siegel, J. Am. Chem. Soc. 113 (1991) 2332;
(c) H.G. Wey, P. Betz, I. Topalovic, H. Butenschon, J. Organomet. Chem. 411
(1991) 369.
[11] (a) M.F. Semmelhack, G.R. Clark, R. Farina, M. Saeman, J. Am. Chem. Soc. 101
(1979) 217;
[19] S. Chamberlin, W.D. Wulff, J. Org. Chem. 59 (1994) 3047.
[20] R.M. Silverstein, G.C. Bassler, T.C. Morrill, Spectrometric Identification of
Organic Compounds, fourth ed., John Wiley and Sons, New York, 1981.
[21] W.D. Wulff, K.-S. Chan, P.-C. Tang, J. Org. Chem. 49 (1984) 2293.
[22] K.H. Dötz, W. Kuhn, K. Ackermann, Z. Naturforsch. 38b (1983) 1351.
[23] S.L.B. Wang, Ph.D. Thesis, The University of Chicago, Chicago, IL, 1991.
[24] W.D. Wulff, W.E. Bauta, R.W. Kaesler, P.J. Lankford, R.A. Miller, C.K. Murray,
D.C. Yang, J. Am. Chem. Soc. 112 (1990) 3642.
(b) E.P. Kundig, C. Grivet, E. Wenger, G. Bernarinelli, A.F. Williams, Helv. Chim.
Acta 74 (1991) 2009.
[12] (a) T.A. Albright, B.K. Carpenter, Inorg. Chem. 19 (1980) 3092;
(b) A. Solladie-Cavallo, G. Wipff, Tetrahedron Lett. (1980) 3047.
[13] (a) W.R. Jackson, I.D. Rae, M.G. Wong, M.F. Semmelhack, J. Garcia, J. Chem. Soc.,
Chem. Commun. (1982) 1359;
(b) W.R. Jackson, I.D. Rae, M.G. Wong, Aust. J. Chem. 39 (1986) 303;
(c) J.C. Boutonnet, L. Mordenti, E. Rose, O. Le Martret, G. Precigoux, J.
Organomet. Chem. 221 (1981) 147;
[25] G.M. Bodner, L.J. Todd, Inorg. Chem. 13 (1974) 360.
[26] F. van Meurs, J.M. van der Toorn, H. van Bekkum, J. Organomet. Chem. 113
(1976) 341.
(d) H. Paramahamsan, J.M. Pearson, A.A. Pinkerton, E.A. Zhurova,
Organometallics 27 (2008) 900.
[27] B.P. Roques, C. Segard, S. Combrisson, F. Wehrli, J. Organomet. Chem. 73 (1974)
327.
[14] (a) B. Ohlsson, C. Ullenius, S. Jagner, C. Grivet, E. Wenger, E.P. Kundig, J.
Organomet. Chem. 365 (1989) 243;
[28] (a) R. Hoffmann, T.A. Albright, D.L. Thorn, Pure Appl. Chem. 50 (1978) 1;
(b) A. Elass, J. Brocard, G. Surpateanu, G. Vergoten, THEOCHEM 466 (1999) 35.
[29] A. Pfletschinger, W. Koch, H.-G. Schmalz, New J. Chem. 25 (2001) 446.
(b) E.P. Kundig, V. Desobry, D.P. Simmons, E. Wenger, J. Am. Chem. Soc. 111
(1989) 1804.