Regioselective formation of β-alkyl-α-phenyliridabenzenes via unsymmetrical 3-vinylcyclopropenes: Probing steric and electronic influences by varying the alkyl ring substituent

Article abstract of DOI:10.1002/chem.200400893

The synthesis of unsymmetrical (Z)-1-alkyl-3-(2-iodovinyl)-2-phenyl-1- cyclopropenes (R = Me (8a), Et (8b), iPr (8c), and tBu (8d)) and their reactions with Vaska's complex [Ir(CO)Cl(PPh₃)₂] and its trimethylphosphine analogue [Ir(CO

Full text of DOI:10.1002/chem.200400893

FULL PAPER
Regioselective Formation of b-Alkyl-a-phenyliridabenzenes via
Unsymmetrical 3-Vinylcyclopropenes: Probing Steric and Electronic
Influences by Varying the Alkyl Ring Substituent**
He-Ping Wu, Timothy J. R. Weakley, and Michael M. Haley*[a]
Dedicated to Professor Armin de Meijere on the occasion of his 65th birthday
Abstract: The synthesis of unsymmetri-
cal (Z)-1-alkyl-3-(2-iodovinyl)-2-
22) were obtained in modest yields and
were fully characterized by spectro-
scopic means. X-ray structural data was
secured for iridabenzvalene 13d and
iridabenzenes 14a,b,d. Whereas irida-
benzenes 14a–c were stable at 758C
for 48 h, 14d, which possesses a bulky
tBu group, rearranged cleanly to cyclo-
pentadienyliridium 15d at 508C over
15 h and displayed first-order kinetics.
The influence of the alkyl substituent
on the mechanisms of iridacycle gener-
ation, isomerization, and iridabenzene
regioselectivity is discussed.
phenyl-1-cyclopropenes (R=Me (8a),
Et (8b), iPr (8c), and tBu (8d)) and
their reactions with Vaskaꢀs complex
[Ir(CO)Cl(PPh3)2] and its trimethyl-
phosphine analogue [Ir(CO)Cl(PMe3)2]
were investigated. Iridabenzvalene (13/
20), iridabenzene (14/21), and/or h5-cy-
clopentadienyliridium complexes (15/
Keywords: aromaticity · iridium ·
metallacycles · strained molecules ·
valence isomerization
Introduction
Metallabenzenes are a rare class of aromatic molecules
where a transition metal fragment (MLn) has replaced one
of the benzene methine (CH) units.[1] Although more than
two dozen such structures have been prepared over the last
20 years or so, most examples have been stabilized by inclu-
sion of a heteroatom (O,N,S) or by h6-coordination to a
second transition metal fragment. Until recently, only two
families of discrete metallabenzenes were known, namely
Roperꢀs osmabenzenes (e.g., 1)[2] and Bleekeꢀs iridabenzenes
(e.g., 2);[3] nonetheless, all investigations into both families
originated from compounds 1 and 2, respectively. Very
recent work on osmabenzenes[4] and irida-aromatic com-
pounds[5] has provided a few additional examples; however,
the generality of these methods has yet to be determined.
Therein lies a key difficulty within this field of research:
lack of versatile synthetic routes has meant that detailed,
fundamental study of metallabenzenes has been prob-
lematic.
We recently reported the preparation of a third family of
metallabenzenes (e.g., 3) using 3-vinyl-1-cyclopropenes (e.g.,
4) as the source of the metallabenzene ring carbon atoms.[6]
Unlike previous syntheses, this new method allows us to ra-
tionally alter variables (metal center, ligands on the metal,
substituents on the C5 backbone) to perform detailed struc-
ture–property relationship investigations. We believe such
augmented studies are now possible since: 1) our route per-
mits direct access into the metallabenzene manifold. There
is no need for subsequent oxidative and/or reductive trans-
formation(s) of the resultant organometallic species. If the
metallabenzene does not form initially, simple heating iso-
[a] Dr. H.-P. Wu, Dr. T. J. R. Weakley, Prof. Dr. M. M. Haley
Department of Chemistry
University of Oregon
Eugene, OR, 97403–1253 (USA)
Fax : (+1)541-346-0487
E-mail: haley@uoregon.edu
[**] Metallabenzenes and Valence Isomers, Part 8. For Part 7 see refer-
ence [9b].
Chem. Eur. J. 2005, 11, 1191 – 1200
DOI: 10.1002/chem.200400893
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1191

merizes the intermediate benzvalene (e.g., 5).[7] 2) This
route permits the use of a variety of starting transition
metal complexes. In addition to iridabenzenes, we have very
recently extended our methodology to examples containing
Rh (6)[8] and Pt (7);[9] thus, this appears to be the first gener-
al pathway to metalla-aromatic formation. 3) Depending
upon the vinylcyclopropene used as the C5 backbone, a vari-
ety of substituted metallacycles,[10] previously inaccessible
substitution patterns, and new metalla-aromatic topologies
should become available. As an extension of our previous
studies, and to elaborate further on point 3, we report
herein the synthesis of several (Z)-1-alkyl-2-phenyl-3-(2-io-
dovinyl)-1-cyclopropenes (e.g., 8) and their reactivity with
Vaskaꢀs complex and its PMe3 analogue, resulting in regiose-
lective iridabenzene formation. We also discuss the isomeri-
zation and kinetic behavior of the resultant iridabenzvalenes
and benzenes, and probe the mechanism of their intercon-
version by fine-tuning the alkyl substituents.
The tert-butyl/phenyl derivative appears to be the upper
limit of the carbene addition route as the analogous reaction
failed to furnish the cyclopropene ester from di-tert-butyl-
acetylene. The very low yield of 9d is somewhat disappoint-
ing as it was hoped that the more reactive [Rh2(OAc)4] cata-
lyst would be superior to CuSO4, the originally utilized cata-
lyst;[13] instead, the yields of isolated material were compara-
ble.
Reduction of esters 9 with excess DIBAL-H[15] afforded
alcohols 11 in 75–95% yield. Unfortunately, use of equimo-
lar amounts of DIBAL-H gave mixtures of the desired alde-
hyde 12 product along with alcohol 11 and unreacted start-
ing material. Oxidation of 11 with the Dess–Martin re-
agent[16] afforded reasonable yields of aldehydes 12 as mod-
erately stable oils. The notable exception was 12a, where
the crude isolated material was immediately subjected to
the Wittig reaction because of its rather high instability. Use
of Ph3P=CHI[17] gave the vinylcyclopropenes 8 in 48–72%
yield with high (>15:1 Z:E) stereoselectivity. Similar to 4,
ligands 8 can be prepared in 750 mg–1 g quantities; howev-
er, the molecules are somewhat unstable and should be
stored in the dark at ꢀ208C with a small amount hydroqui-
none to inhibit decomposition.
Results and Discussion
Ligand synthesis: Preparation of vinylcyclopropenes 8 was
achieved by the route depicted in Scheme 1. By using the
same procedure that previously produced 9a[11] and 9b,[12]
Reactions with Vaskaꢀs complex [Ir(CO)Cl(PPh3)2]: Treat-
ment of ligand 8d with BuLi at ꢀ788C followed by addition
of Vaskaꢀs complex produced a yellow-orange solution upon
warming over 3 h from ꢀ78 to 08C, from which iridabenz-
valene complex 13d could be isolated in 32% yield
(Scheme 2). Purification of 13d was achieved by recrystalli-
Scheme 2.
zation in toluene/cyclohexane. Although stable in the solid
state, solutions of 13d in C6D6 at 208C immediately began
to isomerize to an iridabenzene. After four days at 208C,
Scheme 1.
1
the H NMR spectrum indicated that 13d had disappeared
completely to afford a mixture of iridabenzene 14d and cy-
clopentadienyliridium complex 15d in a ratio of about 3:1 in
94% combined yield. The regiochemistry of 14d as the b-
tert-butyl isomer was confirmed by crystallographic analysis
(vide infra); the a-tert-butyl regioisomer was not detected in
the transformation. We believed initially that 15d was most
likely derived from the thermally unstable a-regioisomer by
ring contraction via “carbene migratory insertion”,[18] a reac-
[Rh2(OAc)4]-catalyzed addition of ethyl diazoacetate to
either isopropyl- or tert-butylphenylacetylene yielded the
[2+1] cycloadducts 9c and 9d,[13] respectively. Interestingly,
the bulky alkyl groups on the starting alkynes result in com-
petitive carbene insertion into the benzene ring, as demon-
strated by the isolation of ethyl 4-isopropyl- or 4-tert-but-
ylethynyl-2,4,6-cycloheptatrien-1-yl carboxylate (10c,d).[14]
1192
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 1191 – 1200

Metallabenzenes and Valence Isomers
FULL PAPER
tion facilitated by the steric hindrance between the bulky
tBu and [Ir(CO)(PPh3)2] moieties. To test this hypothesis,
solutions of both 14d and benzvalene 13d in C6D6, respec-
tively, were prepared and the kinetic conversion processes
monitored by 1H NMR spectroscopy under the same re-
action conditions. The data showed that at 208C 14d re-
arranged into 15d and that the rate of transformation of
14d into 15d was faster than that of 13d to 15d. These ex-
perimental results suggest that the initially formed 15d was
predominantly from the rearrangement of b-tert-butyl re-
gioisomer 14d and not from the corresponding a-isomer;
therefore, the valence isomerization of unsymmetrical benz-
valene 13d to benzene 14d appears to be highly regioselec-
tive. At 508C for 15 h, 14d rearranged quantitatively to 15d
with dissociation of PPh3; however, the rearrangement could
be inhibited by addition of two equivalents of PPh3 prior to
heating. In this latter case, complete transformation to 15d
took about 50 h.
Comparison of the above results with those generated
from the reactions with cyclopropene 4 and the phenyl/tri-
methylsilyl analogue reveal some interesting trends. With
Vaskaꢀs complex, the stability of the resultant iridabenz-
valenes is influenced mainly by electronic donation of the
substituents on the cyclopropyl ring. In the case of 13d, its
stability in solution is intermediate between the diphenyl
(not detected at 208C)[6] and phenyl/trimethylsilyl (16, stable
for several days at 208C)[10] systems, which is in agreement
with intermediate electron donation of tBu between Ph and
SiMe3 groups. Electron donation of the alkyl group increases
Scheme 3.
zene. Benzvalene compounds 13a–c were not isolated from
the reaction mixture because of their relatively rapid isom-
erization to 14a–c, which were isolated in about 25–30%
yield. After 8 h and 30 h at 208C in C6D6 solution, 13a,b
and 13c, respectively, had isomerized completely to give
14a–c as well as very minor amounts of the a-alkyl re-
gioisomer 19a–c, as detected by NMR spectroscopy of the
crude reaction mixtures. Unlike 14d, Ar-blanketed solutions
of 14a–c were stable for over 48 h at 758C. These results
suggest a trend of iridabenzvalene stability as 13a=13b<
13c<13d, which is in agreement with increasing electronic
donation. Somewhat surprising is the independence of the
stability of 13 from alkyl group sterics. On the other hand,
iridabenzene stability is ordered as 14a=14b>14c>14d,
which is now in agreement with increasing alkyl sterics.
Reactions with [Ir(CO)Cl(PMe3)2]: Although PMe3 ligands
are known to stabilize iridabenzvalenes by stronger elec-
tronic donation compared to PPh3,[6b,7] we wanted to exam-
ine how the decreased sterics of the [Ir(CO)(PMe3)2] frag-
ment influenced both the regioselectivity of isomerization of
unsymmetrical benzvalene compounds and the stability of
the resultant benzene compounds. Reaction of cyclopro-
penes 8b,d with [Ir(CO)Cl(PMe3)2] at ꢀ78 to 08C over 3 h
yielded the corresponding benzvalenes 20b,d as the only
product (Scheme 4). Both complexes were purified by re-
crystallization in hexane at ꢀ308C and are stable at 208C.
Isomerization was observed at higher temperatures, with
20b completely converting to b-ethyl regioisomer 21b at
the electron density of cyclopropene double bond, which in
turn strengthens the h2-cyclopropene interaction with the iri-
dium center. On the other hand, the contiguous arrange-
ment of the iridium fragment and neighboring phenyl group
with the bulky tBu and SiMe3 substituents destabilizes the
resultant iridabenzene compounds. Whereas solutions of 3
are stable at 1008C over 24 h,[6] and the b-trimethylsilyl ben-
zene 17 (stable at 208C) rearranges to cyclopentadienyl
complex 18 at 758C over 24 h,[10] 14d converts to 15d even
at 208C. This difference between 14 and 17 is likely due to
closer proximity of the tBu group, a result of the shorter
ꢀ
C C bond compared to a C-Si bond (1.52 versus 1.85 ꢂ, re-
spectively).
Reactions of Vaskaꢀs complex with cyclopropenes 8a–c
showed the influence of alkyl substituent on both the forma-
tion of 13 and its transformation to 14 (Scheme 3). All three
reactions yielded yellow solutions below 08C, which then
1
turned red-brown upon warming to 208C. H NMR spectro-
scopy showed that each crude reaction mixture was com-
posed of the corresponding iridabenzvalene and iridaben-
Scheme 4.
Chem. Eur. J. 2005, 11, 1191 – 1200
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1193

M. M. Haley et al.
758C over 4 h. Under the same conditions, however, tBu an-
alogue 20d gave a mixture composed of unreacted benzva-
lene, benzene 21d, and cyclopentadienyliridium complex
22d in a ratio of 5:10:1, as well as partial decomposition.
Prolonged heating of this mixture led to complete decompo-
sition affording unidentified materials. Interestingly, we
never observe evidence for formation of the a-regioisomer
of 21, even in the NMR spectra of the crude reaction mix-
tures; thus, the lower sterics of PMe3 than PPh3 does not de-
crease the regioselectivity of the isomerization. The influ-
ence of the Et and tBu substituents on the stability and iso-
merization rate of 20 is similar to that observed for corre-
sponding PPh3 analogs 13. Additionally, the higher stability
of the b-tBu substituted 21d compared to the corresponding
PPh3 analogue 14d indicates that the PMe3 ligand does sta-
bilize the iridabenzene. Nevertheless, we have no definitive
conclusions on this steric/electronic influence.
Figure 1. Molecular structure of iridabenzvalene 13d. The thermal ellip-
soids are drawn at the 30% probability level. Only the ipso-PPh3 carbon
atoms are shown for clarity. Selected bond lengths [ꢂ] and bond angles
1
Iridabenzvalene characterization: In the H NMR spectra of
13 and 20, a characteristic broad signal around d=3.2 ppm
is assigned to the proton resonance of the sp3-CH fragment
(H3). Two complex multiplets in the ranges of d=6.05–6.45
(H4) and 6.72–7.17 ppm (H5) are attributed to the two sp2-
CH groups of the bridging double bond, with the latter
signal being the CH attached to the metal center. The
13C NMR spectra of the benzvalenes show significant upfield
shifts of the resonances for the coordinated cyclopropene
double bond atoms C1 and C2, compared to the correspond-
ing uncoordinated carbon atoms, which is commonly ob-
served in olefin h2-metal complexes. For example, in 20b the
resonances for C1/2 are at d=65.88 and 71.03 ppm, versus
d=110.89 and 117.90 ppm, respectively, in 8b. The lack of
symmetry in the benzvalenes leads to a more complicated
pattern of C–P couplings than in our previous work. In 20b,
coupling of the two cis-PMe3 to C6 in the CO ligand and to
C5 results in two triplets with coupling constants of 7.6 Hz
and 15.1 Hz, respectively. Coupling of the cis- and trans-
PMe3 to C1 and C2 furnish two doublet of doublet signals
with coupling constants of 72.5/4.0 and 70.5/3.0 Hz, respec-
tively. The lack of symmetry is also exhibited in the
31P NMR spectrum by the appearance of two doublets at
d=ꢀ0.09 and ꢀ3.92 ppm. The absorption band of the CO
group in 20b was observed at 1966 cmꢀ1 in the IR spectrum.
The structure of benzvalene 13d was further confirmed by
single crystal X-ray diffraction. Selected bond lengths and
angles are given in Figure 1. The Ir atom of 13d has a tor-
sion trigonal-bipyramidal coordination configuration com-
posed of carbonyl, h2-cyclopropene double bond, two phos-
phine, and s-vinyl ligands around the Ir atom. Comparison
of key bond distances and bond angles with data for known
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
[8]: Ir P 2.353(3), Ir C1 2.19(2), Ir C2 2.22(2), Ir C5 2.18(3), Ir C6
ꢀ
ꢀ
ꢀ
ꢀ
1.73(2), C1 C2 1.41(3), C1 C3 1.56(3), C1 C7 1.47(3), C2 C3 1.57(3),
ꢀ
ꢀ
ꢀ
C2 C13 1.49(3), C3 C4 1.40(3), C4 C5 1.38(4); P-Ir-Pi 110.4(1), P-Ir-C2
136.4(5), Pi-Ir-C1 146.6(5), C1-Ir-C2 37.1 (6), C5-Ir-C6 174(1), C1-Ir-C5
81.5(8), C2-Ir-C5 80.4(8), C3-C1-C7 130(2), C3-C2-C13 129(2), C1-C3-C2
54(1), C3-C4-C5 115(2), C4-C5-Ir 111(2).
Table 1. Selected X-ray bond lengths [ꢂ] and bond angles [8] for irida-
benzvalenes 5, 13d, and 23.
5
23
13d
ꢀ
Ir C1
ꢀ
Ir C2
2.146
2.143
2.095
1.447
39.4
178.9
104.1
116.3
2.172
2.159
2.092
1.440
38.8
177.2
109.4
116.2
2.189
2.220
2.184
1.405
37.1
174.0
110.4
109.9
ꢀ
Ir C5
ꢀ
C1 C2
C1-Ir-C2
C5-Ir-C6
P-Ir-P
dihedral
weaker bonding is corroborated by the fact that 13d isomer-
izes at room temperature to 14d, whereas benzvalene com-
pounds 5 and 23 similarly are stable (vide supra).
Iridabenzene characterization: The spectroscopic data of 14
and 21 resemble those for known iridabenzenes 3 and 17
with respect to chemical shifts and signal patterns.[6,10] The
low-field pseudo-quartet in the characteristic range of d=
10.5–11.0 ppm is assigned to H5, the proton ortho- to the Ir
center. This dramatic downfield shift for H5 can be attribut-
ed mainly to the magnetic anisotropic effect of the heavy
metal on the neighboring proton, an effect that attenuates
benzvalenes 5[7] and 23[6b] (Table 1) shows that the Ir C1
and Ir C2 bond lengths of 13d (2.189, 2.220 ꢂ) are consid-
ꢀ
ꢀ
erably longer than that in its analogues. Furthermore, the
ꢀ
shorter C1 C2 bond length and smaller C1-Ir-C2 bond
angle and Ir-C1-C2-C3 dihedral all suggest that the h2-inter-
action of the cyclopropene p-bond with the Ir center in 13d
is weaker than in the analogues shown in Table 1. This
1194
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 1191 – 1200

Metallabenzenes and Valence Isomers
FULL PAPER
Table 2. Selected NMR chemical shift data for iridabenzenes.[a]
Mechanisms for iridabenzva-
lene formation/isomerization:
The proposed mechanisms for
Cmpd
H3
H4
H5
C1
C2
C3
C4
C5
C6
14a
14b
14c
14d
3
21b
21d
24
8.30
8.29
8.41
8.69
8.44
8.13
8.46
8.25
7.75
7.79
7.86
7.83
7.79
7.90
7.88
7.91
10.62
10.61
10.60
10.54
10.79
10.92
10.87
11.00
187.46
187.31
187.65
186.68
187.60
188.83
140.47
140.84
145.52
145.61
141.93
138.55
133.93
139.99
–
138.89
139.99
135.38
129.07
129.48
129.85
129.63
127.76
129.70
183.92
184.95
185.52
186.55
187.43
174.20
204.97
204.17 iridabenzvalene isomerization/
[b]
203.90
208.32
201.09
181.02
benzene formation have been
discussed at length previous-
ly;[6b,7] thus, only salient points
[c]
[c]
[c]
[c]
[c]
[c]
will be presented herein. Benz-
valene formation continues to
appear to be influenced prefer-
entially by electronic factors.
–
–
–
–
–
–
189.95
141.49
136.74
129.06
176.04
189.44
1
1
[a] All data acquired in C6D6. Assignments based on 2D H–1H COSY and 2D H–13C HMQC and HMBC ex-
periments. Atom labeling as shown in Figure 2. [b] Resonance obscured. [c] Inseparable mixture of 20d, 21d,
and 22d and also decomposition products precluded accurate assignment of resonances.
significantly with increasing distance.[19] The para- and meta-
protons on the iridabenzene rings resonate in the range of
d=8.7–8.1 ppm and d=7.9–7.7 ppm, respectively, values
that are closer to normal resonances caused mainly by aro-
matic ring currents. A summary of the NMR chemical shifts
for the ring proton/carbon atoms in both iridabenzenes 14
and 21 as well as comparison with related systems 3 and 24
are given in Table 2. Replacement of PPh3 with PMe3 on the
iridium center leads to an upfield shift of the C3 (Dd=
5.5 ppm), C5 (Dd=10.8 ppm), and C6 (Dd=23.2 ppm) atom
resonances, respectively, which can be rationalized in the
terms of electronic influences. The stronger donating ability
of PMe3 compared to PPh3 results in a more electron-rich
metal center, which in turn increases the electron density of
benzene ring and thus the upfield carbon shifts. On the
Figure 2. Molecular structure of iridabenzene 14d. The thermal ellipsoids
other hand, small shifts for C1 (Dd=1.5 ppm), C2 (Dd=
1.4 ppm), and C4 (Dd=0.2 ppm) were observed, indicating
the lack of sensitivity of these carbon atoms. The 31P NMR
spectra show a single resonance for the two phosphine li-
gands in the region of d=18.5 ppm (PPh3) and d=
ꢀ38.6 ppm (PMe3), respectively, due to rapid exchange be-
tween axial and basal phosphine ligands in solution at
208C.[3,20] The vibration frequency of CO group in iridaben-
zenes falls in the range between 1892–1982 cmꢀ1.
are drawn at the 30% probability level. Selected bond lengths [ꢂ] and
ꢀ
ꢀ
ꢀ
ꢀ
bond angles [8]: Ir P1 2.348(2), Ir P2 2.306(3), Ir C1 2.054(9), Ir C5
ꢀ
ꢀ
ꢀ
ꢀ
2.000(9), Ir C6 1.911(11), C1 C2 1.407(12), C2 C3 1.382(13), C3 C4
ꢀ
ꢀ
ꢀ
1.427(12), C4 C5 1.343(12), C1 C7 1.490(12), C2 C13 1.554(13); P1-Ir-
P2 105.61(9), P1-Ir-C1 137.5(2), P2-Ir-C1 116.5(2), C5-Ir-C6 172.1(4), C1-
Ir-C5 87.4(4), C1-Ir-C6 91.4(4), Ir-C1-C2 130.8(7), Ir-C1-C7 109.6(6), C1-
C2-C3 119.5(9), C1-C2-C13 125.7(8), C2-C3-C4 128.2(9), C3-C4-C5
123.9(8), Ir-C5-C4 130.2(7), Ir-C6-O 174.1(8).
Confirmation of the structures of iridabenzenes 14a,b,d
was provided by single-crystal X-ray diffraction. The molec-
ular structure of 14d is shown in Figure 2 along with key
bond lengths and angles. Comparison of the bond lengths
and angles of complexes 14a,b,d with analogues 3 and 25 is
given in Table 3. The X-ray analyses verify that the alkyl
groups on all three metallacycles are indeed at the b-posi-
tion to the Ir center, in agreement with the spectroscopic as-
Table 3. Selected X-ray bond lengths [ꢂ] and bond angles [8] for irida-
benzenes 3, 14a, 14b, 14d, and 25.
14a
14b
14d
3
25
ꢀ
Ir C1
ꢀ
Ir C5
2.020
2.012
1.423
1.335
1.382
1.408
88.5
2.029
2.023
1.413
1.390
1.393
1.360
87.0
2.054
2.000
1.407
1.382
1.427
1.343
87.4
2.021
2.025
1.409
1.410
1.377
1.334
86.9
2.047
2.004
1.427
1.372
1.386
1.381
88.1
ꢀ
C1 C2
ꢀ
C2 C3
ꢀ
C3 C4
ꢀ
C4 C5
ꢀ
signment. The C C bond lengths in the metallacycles are es-
C1-Ir-C5
sentially equal and thus can be regarded as evidence of de-
mean deviation
0.011
0.008
0.009
0.024
0.041
ꢀ
localization of the aromatic ring p electrons. The mean C C
bond lengths are extremely close—1.387 ꢂ (14a), 1.389 ꢂ
ꢀ
ꢀ
(14b), and 1.386 ꢂ (14d). The Ir C1 and Ir C5 bond
lengths in 14a,b,d average around 2.02 ꢂ and are intermedi-
As mentioned above, the trend of iridabenzvalene stability
for the PPh3-containing cycles (13a=13b<13c<13d<16)
is in agreement with increasing electronic donation of the
cyclopropenyl substituents yet is seemingly independent of
their steric effects.
ꢀ
ate between Ir C single and double bonds. The ring of
every iridabenzene is almost planar with mean deviations of
0.011 ꢂ (14a), 0.008 ꢂ (14b), and 0.009 ꢂ (14d). Contrary
to our expectations based on reactivity, the bulky tBu group
does not induce greater torsion strain in the iridabenzene
ring of 14d, at least not in the solid state.
The “valence isomerization” of 13 to 14 still presents sev-
eral problems. Although a concerted process (path 1) is the
Chem. Eur. J. 2005, 11, 1191 – 1200
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1195

M. M. Haley et al.
simplest explanation, and appears to be supported computa-
tionally,[21] this does not readily account for the strong pref-
erence for the b-regioisomer. If the isomerization were
slightly less concerted, involving nucleophilic attack of the
cyclopropene p-bond on the electron-deficient Ir center
(path 2), this presumably should give favorable benzylic car-
bonium ion 26 (Scheme 5). Rearrangement of this stabilized
(18) at 758C.[10] Iridabenzene 14d provided us with an op-
portunity to study the kinetics of this transformation. NMR
spectroscopy was utilized to examine the kinetics of the re-
arrangement of 14d at 313, 323, and 333 K, respectively
(Figure 3). The relative concentrations of 14d and 15d in
Figure 3. Time-evolved 1H NMR spectra of the thermally rearrangement
of 14d to 15d at 323 K in the C6D6. The bottom spectrum was recorded
after heating for 1 h. The following spectra represent 2 h intervals with
top spectrum corresponding to 27 h reaction time. For clarity, the signals
from 0 to 4 ppm were omitted.
Scheme 5.
the course of the reaction were calculated based on the inte-
gration of their ring proton resonances in the H NMR spec-
1
tra. The rate constants of the reaction were determined by
plotting ln[C]i/[C]o versus time. In every case, the data was
linear, indicating first-order kinetics. The slopes of the lines
were taken as the rate constants and were (1.76ꢁ0.06)ꢃ
10ꢀ5, (1.04ꢁ0.02)ꢃ10ꢀ4, and (3.76ꢁ0.16)ꢃ10ꢀ4 sꢀ1 at 313,
323, and 333 K, respectively. An Erying plot of the data
from the rate measurements was linear and gave the follow-
ion, however, would lead predominantly to 19, that is, the
wrong regioisomer. The remaining possibility (path 3) en-
tails reversible dissociation of the h2-cyclopropene in 13 to
give s-vinyliridium intermediate 27 (initially formed by nu-
cleophilic substitution of Vaskaꢀs complex with the vinyl
lithiate). Regioselective insertion of the iridium atom into
¼
¼
the less-electron-rich C C(Ph) s bond leads to intermediate
ing activation parameters: DH =31.2ꢁ1.9 kcalmolꢀ1, DS
ꢀ
Dewar benzene 28, which then affords iridabenzene 14 via
rapid valence isomerization (relief of strain). In addition to
these electronic factors, steric effects also appear to support
path 3. Strong steric repulsion between the bulky iPr/tBu
groups and the [Ir(CO)(PPh3)2] fragment results in a “less
=9.3ꢁ36 calKꢀ1 molꢀ1. These numbers are in good agree-
ment with a recent computational study on the conversion
of metallabenzene complexes to cyclopentadienyl-metal
complexes.[21b,24]
The plausible mechanism for this transformation is shown
ꢀ
efficient approach” of the iridium center to the C C(R)
in Scheme 6. Carbene migratory insertion[18,25] of 14d leads
bond in 27, thus the observed decrease in the rate of valence
isomerization. Detracting from this latter pathway has been
our inability to isolate or even detect spectroscopically an
irida-Dewar benzene complex such as 28. It should also be
noted that more than one of these pathways is possibly op-
erational. Regardless of the exact mechanism(s) involved,
isomerization of 13 to 14 and/or 15 is generally a clean,
high-yield process.
Scheme 6.
Mechanism and kinetics for iridabenzene isomerization:
Much more certain mechanistically is the conversion of met-
allabenzenes to cyclopentadienyl-metal complexes. Metalla-
benzenes have been implicated in a number of transforma-
tions which yielded Cp-metal complexes.[9c,22] Jones and Alli-
son have observed the conversion of a transient ruthenaben-
zene to the corresponding Cp2Ru complex using NMR spec-
troscopy at low temperature.[23] Our group recently reported
the rearrangement of two trimethylsilyl-substituted iridaben-
zene isomers into the same cyclopentadienyliridium complex
to formation of coordinatively unsaturated h1-cyclopenta-
dienyl intermediate 29. Subsequent loss of PPh3 yields ener-
getically more stable h5-cyclopentadienyliridium complex
15d. It is reasonable to assume that the interaction of the
contiguous tert-butyl substituent, phenyl group, and metal
fragment may cause some degree of torsion in the metalla-
benzene ring in solution and thus lead to the observed car-
bene migration in 14d and 21d.
1196
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 1191 – 1200

Metallabenzenes and Valence Isomers
FULL PAPER
for 3 h. The solution was then filtered through a small column of silica
gel to remove the catalyst. Concentration of the filtrate and purification
of the crude material by column chromatography on silica gel (hexanes/
Et2O, 6:1) afforded compounds 9 and 10 as colorless oils, with 9 eluting
first.
Conclusion
The reactions of unsymmetrical (Z)-1-alkyl-3-(2-iodovinyl)-
2-phenyl-1-cyclopropenes 8a–d with Vaskaꢀs complex pro-
duced initially iridabenzvalenes 13a–d. Subsequent isomeri-
zation in a highly regioselective manner furnished a-phenyl-
b-alkyl-iridabenzene complexes 14a–d, as well as trace
amounts of the b-alkyl-a-phenyl isomers 19a–c. The a-
phenyl-b-alkyl arrangement of the substituents was con-
firmed by X-ray crystallography for three of the iridaben-
zenes. Use of [Ir(CO)Cl(PMe3)2] afforded benzvalene com-
plexes 20b/d and benzenes 21b/d. The trend of the isomeri-
zation rate of alkyl-substituted benzvalene to benzene com-
plexes is 13a=13b>13c>13d>20b>20d that is in agree-
ment with increase of both electronic donation and sterics
of the alkyl group as well as electron donation and/or re-
duced sterics of the phosphine ligand. Whereas 14a–c and
20b were stable at 758C for 48 h, iridacycle 14d possessing a
bulky tBu group was unstable and rearranged cleanly via
carbene migratory insertion to cyclopentadienyliridium com-
plex 15d at 508C over 15 h and displayed first-order kinet-
ics. Iridabenzene stability was ordered as 21b > 14a=
14b=14c> 21d > 14d in agreement with a decrease of
sterics of the alkyl group and as well as electron donation
and/or less sterics of the phosphine ligand. Future work will
focus on the preparation of irida-aromatics with different
substituents, as well as reaction of cyclopropene ligands 8a–
d with additional transition-metal complexes. The results of
these studies will be reported shortly.
9c: 49%; 1H NMR (CDCl3): d=7.50–7.28 (m, 5H), 4.24–4.06 (m, 2H),
2.98 (sept, J=6.9 Hz, 1H), 2.45 (s, 1H), 1.32 (d, J=6.9 Hz, 3H), 1.31 (d,
J=6.9 Hz, 3H), 1.25 ppm (t, J=7.2 Hz, 3H); 13C NMR (CDCl3): d=
175.92 (CO), 129.41, 128.58, 128.53, 126.95, 115.17, 103.59, 60.00, 26.22,
21.87, 20.67, 14.35 ppm; IR (Et2O):: n˜ =1731 (CO) cmꢀ1
.
10c: 6%; 1H NMR (CDCl3): d=6.81 (d, J=6.4 Hz, 1H), 6.25–6.18 (m,
2H), 5.48–5.35 (m, 2H), 4.24 (q, J=7.2 Hz, 2H), 2.72 (sept, J=6.9 Hz,
1H), 2.58 (t, J=5.6 Hz, 1H), 1.29(t, J=7.2 Hz, 3H), 1.21 ppm (d, J=
6.9 Hz, 6H); 13C NMR (CDCl3): d=172.59 (CO), 134.35, 127.84, 125.96,
125.41, 116.73, 115.72, 96.66, 80.62, 61.01, 43.18, 22.86, 21.03, 14.12 ppm;
IR (Et2O): n˜ =1730 (CO) cmꢀ1
.
9d: 4%; 1H NMR (CDCl3): d=7.51–7.29 (m, 5H), 4.22–4.07 (m, 2H),
2.45 (s, 1H), 1.33 (s, 9H), 1.23 ppm (t, J=7.2 Hz, 3H); 13C NMR
(CDCl3): d=175.91(CO), 129.55, 128.60, 128.52, 126.99, 117.93, 102.35,
59.96, 31.85, 28.36, 21.78, 14.37 ppm; IR (Et2O): n˜ =1744 (CO) cmꢀ1
.
10d: 9%; 1H NMR (CDCl3): d=6.82 (d, J=6.2 Hz, 1H), 6.25–6.19 (m,
2H), 5.48–5.35 (m, 2H), 4.26 (q, J=7.2 Hz, 2H), 2.60 (t, J=5.8 Hz, 1H),
1.31 (t, J=7.2 Hz, 3H), 1.28 ppm (s, 9H); 13C NMR (CDCl3): d=172.73
(CO), 134.39, 128.00, 126.06, 125.48, 116.73, 115.70, 99.48, 79.96, 61.09,
43.26, 30.94, 27.91, 14.18 ppm; IR (Et2O) n˜ = 1742 (CO) cmꢀ1
.
1-Alkyl-3-hydroxymethyl-2-phenyl-1-cyclopropene (11): General proce-
dure: To a mixture of ester 9 (5.0 mmol) and dry THF (30 mL) cooled to
08C was added DIBAL-H (11 mL, 1m in hexane, 11 mmol) by syringe
over 5 min. After stirring for 3 h at 08C, the solution was transferred to a
separatory funnel containing potassium sodium tartrate (Rochelleꢀs salt,
2.0 g) dissolved in H2O (10 mL). After shaking, the gel formed was ex-
tracted with Et2O (3ꢃ25 mL). The combined organic phase was dried
(MgSO4) and concentrated. Column chromatography over silica gel (hex-
anes/Et2O, 3:1) gave 11 as a colorless oil.
11a: 75%; 1H NMR (CDCl3): d=7.51 (d, J=7.9 Hz, 2H), 7.40 (t, J=
7.5 Hz, 2H), 7.30 (t, J=7.3 Hz, 1H), 3.72 (d, J=4.6 Hz, 2H), 2.54 (s),
2.02 (t, J=4.6 Hz, 1H), 1.55 ppm (br, 1H); 13C NMR (CDCl3): d=
129.65, 128.82, 128.54, 127.76, 114.86, 112.98, 67.96, 22.79, 11.77 ppm; IR
Experimental Section
(Et2O) n 3496 (OH) cmꢀ1
.
3-Methyl-1-phenyl-1-butyne,[26] 3,3-dimethyl-1-phenyl-1-butyne,[27] 1,1,1-
triacetoxy-1,1-dihydro-1,2-benziodoxol-3(1H)-one,[16] iodomethyltriphe-
nylphosphonium iodide,[17] and [Ir(CO)Cl(PMe3)2][28] were prepared ac-
cording to the literature. All other compounds were purchased from com-
mercial suppliers and used as received. Column chromatography was per-
formed on Whatman reagent grade silica gel (230–400 mesh). Manipula-
tion of organometallic reagents was carried out using either a Vacuum
Atmospheres inert atmosphere glove box or standard Schlenk techniques.
THF, Et2O, and hexanes were distilled from Na/benzophenone and C6D6
distilled from LiAlH4. All dried solvents were degassed by three freeze/
pump/thaw cycles prior to use. NMR spectra were recorded on a Varian
Unity-INOVA 300 spectrometer at ambient temperature. 1 H, 13C, and
31P NMR spectra were acquired at 299.95, 75.43, and 121.42 MHz, respec-
tively. Chemical shifts for 1H and 13C NMR spectra are reported in parts
per million (d) downfield from tetramethylsilane using the residual sol-
vent signal (CDCl3: 1H 7.26, 13C 77.00; C6D6: 1H 7.16, 13C 128.39) as an
internal standard. The 31P NMR spectrum is referenced relative to exter-
nal H3PO4 or PPh3. Coupling constants are reported in hertz. FT-IR spec-
tra were recorded using a Nicolet Magna 550 FT-IR spectrometer. Melt-
ing points were determined by using a Mel-Temp II capillary melting
point apparatus equipped with a thermocouple and digital thermometer.
Elemental analyses were performed by Robertson Microlit Laboratories,
Inc.
11b: 77%; 1H NMR (CDCl3): d=7.55 (d, J=7.9 Hz, 2H), 7.42 (t, J=
7.4 Hz, 2H), 7.31 (t, J=7.3 Hz, 1H), 3.78–3.71 (m, 2H), 2.72 (q, J=
7.5 Hz, 2H), 2.06 (t, J=4.7 Hz, 1H), 1.43 (br, 1H), 1.36 ppm (t, J=
7.5 Hz, 3H); 13C NMR (CDCl3): d=129.54, 128.96, 128.56, 127.82, 120.22,
112.31, 68.29, 22.61, 20.14, 12.60 ppm; IR (Et2O): n˜ = 3491 (OH) cmꢀ1
.
11c: 95%; 1H NMR (CDCl3): d=7.56 (d, J=7.9 Hz, 2H), 7.40 (t, J=
7.5 Hz, 2H), 7.30 (t, J=7.4 Hz, 1H), 3.76 (dd, J=10.7, 4.7 Hz, 1H), 3.66
(dd, J=10.7, 4.7 Hz, 1H), 2.99 (sept, J=6.8 Hz, 1H), 2.05 (t, J=4.7 Hz,
1H), 1.50 (br, 1H, OH), 1.33 (d, J=6.8 Hz, 3H), 1.31 ppm (d, J=6.8 Hz,
3H); 13C NMR (CDCl3): d=129.47, 129.14, 128.55, 127.82, 124.07, 111.46,
68.55, 26.86, 22.38, 21.25, 21.17 ppm; IR (Et2O): n˜ = 3493 (OH) cmꢀ1
.
11d: 93%; 1H NMR (CDCl3): d=7.55 (d, J=7.9 Hz, 2H), 7.39 (t, J=
7.5 Hz, 2H), 7.28 (t, J=7.3 Hz, 1H), 3.78 (dd, J=10.5, 4.7 Hz, 1H), 3.62
(dd, J=10.5, 4.7 Hz, 1H), 2.03 (t, J=4.7 Hz, 1H), 1.58 (br s, 1H),
1.31 ppm (s, 9H); 13C NMR (CDCl3): d=129.45, 129.36, 128.60, 127.86,
126.87, 110.08, 68.72, 31.86, 28.86, 22.31 ppm; IR (Et2O): n˜ = 3497 (OH)
cmꢀ1
.
2-Alkyl-3-phenyl-2-cyclopropene-1-carboxaldehyde (12): General proce-
dure: A solution of alcohol 11 (5.0 mmol) in CH2Cl2 (15 mL) was added
dropwise to a mixture of 1,1,1-triacetoxy-1,1-dihydro-1,2-benziodoxol-
3(1H)-one (2.8 g, 7.0 mmol) and CH2Cl2 (30 mL). After stirring for
30 min at 208C, the reaction mixture was treated with 1n NaOH
(50 mL). The mixture was extracted with Et2O (2ꢃ50 mL) and the organ-
ic phase washed with saturated NaCl solution, dried (MgSO4), and con-
centrated. Column chromatography over silica gel (hexanes/Et2O, 5:1)
furnished 12 as a colorless, moderately stable oil. The exception was 12a,
where the crude isolated material was immediately subjected to the
Wittig reaction because of its rather high instability.
Ethyl 2-alkyl-3-phenyl-2-cyclopropene-1-carboxylate (9) and Ethyl 4-al-
kylethynyl-2,4,6-cycloheptatriene-1-carboxylate (10): General prece-
dure:[12] To a solution of alkyne (50 mmol) and [Rh2(OAc)4] (398 mg,
0.9 mmol) in CH2Cl2 (20 mL) was added a mixture of ethyl diazoacetate
(4.0 g, 35 mmol) in CH2Cl2 (5 mL) at a rate of 0.4 mLhꢀ1 by means of a
syringe pump. Upon completion of the addition, stirring was continued
Chem. Eur. J. 2005, 11, 1191 – 1200
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1197

M. M. Haley et al.
12a: 60%; 1H NMR (CDCl3): d=8.87 (d, J=7.3 Hz, 1H), 7.27–6.99 (m,
5H), 2.61 (d, J=7.3 Hz, 1H), 2.36 ppm (s, 3H).
5.2 Hz), 149.50 (dd, J=15.5, 12.5 Hz), 144.09 (d, J=4.0 Hz), 143.12 (t, J=
8.1 Hz), 136.85 (d, J=43.3 Hz), 134.81 (d, J=43.3 Hz), 133.96 (d, J=
11.1 Hz), 133.21 (d, J=11.1 Hz), 128.62 (d, J=1.9 Hz), 128.24 (d, J=
1.8 Hz), 126.79 (s), 126.77 (d, J=10.1 Hz), 126.63 (d, J=9.1 Hz), 122.62
(s), 81.32 (dd, J=71.4, 6.0 Hz), 65.26 (dd, J=62.2, 4.5 Hz), 54.72 (t, J=
3.7 Hz), 35.24 (t, J=4.0 Hz), 29.46 ppm (d, J=2.0 Hz) (2 aromatic reso-
nances obscured); 31P NMR (C6D6): d=ꢀ0.09 (d, J=28.1 Hz),
ꢀ3.92 ppm (d, J=28.1 Hz); IR (CH2Cl2): n˜ = 1971 (CO) cmꢀ1; elemental
analysis calcd (%) for C52H47IrOP2: C 66.29, H 5.03; found: C 65.97, H
4.77.
12b: 61%; 1H NMR (C6D6): d=8.97 (d, J=7.3 Hz, 1H), 7.27–6.98 (m,
5H), 2.61 (d, J=7.3 Hz, 1H), 2.10 (dq, J=7.5, 2.6 Hz, 2H), 0.86 ppm (t,
J=7.5 Hz, 3H); 13C NMR (C6D6): d=203.51, 130.82, 129.60, 129.04,
128.91, 112.24, 105.51, 35.24, 19.76, 11.89 ppm; IR (Et2O): n˜ = 1707 (CO)
cmꢀ1
.
12c: 46%; 1H NMR (CDCl3): d=8.89 (d, J=7.6 Hz, 1H), 7.53–7.37 (m,
5H), 3.05 (sept, J=6.8 Hz, 1H), 2.64 (d, J=7.6 Hz, 1H), 1.36 (d, J=
6.8 Hz, 3H), 1.32 ppm (d, J=6.8 Hz, 3H); 13C NMR (CDCl3): d=205.68,
129.49, 129.18, 128.78, 127.02, 115.62, 104.01, 35.13, 26.64, 20.67,
Iridabenzene 14d and cyclopentadienyliridium 15d: Benzvalene 13d
(188 mg, 0.2 mmol) was dissolved in benzene (5 mL) and kept at 208C
over four days until 11d had completely disappeared. The resultant
brown solution was concentrated and the residue redissolved in hexane/
Et2O (1:1; 10 mL) which was then stored overnight at ꢀ308C. The solid
was collected and washed with cold Et2O to give 14d (130 mg, 71%) as
20.63 ppm; IR (Et2O): n˜ = 1708 (CO) cmꢀ1
.
12d: 68%; 1H NMR (CDCl3): d=8.89 (d, J=7.6 Hz, 1H), 7.56–7.39 (m,
5H), 2.67 (d, J=7.6 Hz, 1H), 1.37 ppm (s, 9H); 13C NMR (CDCl3): d=
205.62, 129.64, 129.19, 128.82, 127.00, 118.45, 102.71, 35.09, 31.73,
28.26 ppm; IR (Et2O): n˜ = 1706 (CO) cmꢀ1
.
1
red crystals. 14d: H NMR (C6D6): d=10.54 (q, J=11.1 Hz, 1H), 8.69 (q,
(Z)-1-Alkyl-3-(2-iodoethenyl)-2-phenyl-1-cyclopropene (8): General pro-
cedure: Iodomethyltriphenylphosphonium iodide was dried under
vacuum at 1008C for 30 min prior to use. The dry salt (1.6 g, 3.0 mmol)
was dissolved in THF (10 mL) and NaN(SiMe3)2 (3 mL, 1.0m in THF,
3.0 mmol) was added by syringe. The reaction was stirred for 5 min,
giving a yellow solution, and then it was cooled to ꢀ788C. To the cold re-
action was successively added HMPA (2.0 mL) and a solution of alde-
hyde 12 (3.0 mmol) in THF (5 mL). The reaction was warmed to ꢀ308C
and stirred for 45 min. Et2O (30 mL) was added to the reaction, the resul-
tant suspension filtered, and the filter cake washed with additional Et2O.
The filtrate was washed with 10% HCl solution, saturated NaCl solution,
and 10% Na2CO3 solution. The organic phase was dried (MgSO4), fil-
tered, and concentrated to give an light brown semi-solid. Chromatogra-
phy on silica gel (hexanes) gave 8 as a colorless oil. The vinyl iodide was
stored at ꢀ308C with a small amount (~1 mg) of hydroquinone to pre-
vent decomposition.
J=6.8 Hz, 1H), 7.84 (dd, J=9.6, 8.3 Hz, 1H), 7.36 (d, J=7.2 Hz, 2H),
7.20–7.15 (m, 12H), 7.10 (t, J=7.7 Hz, 2H), 6.94–6.89 (m, 18H), 6.85 (t,
J=7.4 Hz, 1H), 1.58 ppm (s, 9H); 13C NMR (C6D6): d=208.32 (t, J=
51.9 Hz), 186.68 (t, J=6.1 Hz), 186.55 (s), 172.21 (t, J=3.4 Hz), 145.61 (t,
J=5.5 Hz), 138.89 (s), 136.68–136.24 (m), 134.42 (t, J=5.4 Hz), 129.63 (t,
J=4.2 Hz), 129.48 (s), 125.21 (s), 124.16 (t, J=6.7 Hz), 122.94 (s), 39.76
(t, J=3.0 Hz), 35.77 ppm (s) (1 aromatic resonance obscured); 31P NMR
(C6D6): d=18.24 ppm (s); IR (CH2Cl2): n˜ = 1981 (CO) cmꢀ1; elemental
analysis calcd (%) for C52H47IrOP2: C 66.29, H 5.03; found: C 66.09, H
4.92.
The filtrate from above was concentrated and the residue purified by
chromatography on silica gel (hexanes/Et2O, 5:1) to give compound 15d
(32 mg, 23%) as a yellow solid. 15d: 1H NMR (C6D6): d=7.88–7.78 (m,
8H), 7.14–6.96 (m, 12H), 5.18 (t, J=2.2 Hz, 1H), 4.76 (t, J=2.2 Hz, 1H),
4.18 (dt, J=2.7, 1.1 Hz, 1H), 1.34 ppm (s, 9H); 13C NMR (C6D6): d=
179.83 (d, J=17.1 Hz), 138.12 (s), 137.12 (d, J=35.3 Hz), 134.44 (s),
134.29 (d, J=12.1 Hz), 129.92 (d, J=3.0 Hz), 127.99 (d, J=11.1 Hz),
127.81 (s), 127.27 (s), 115.37 (d, J=7.1 Hz), 108.61 (d, J=6.4 Hz), 88.26
(s), 82.07 (s), 79.56 (s), 32.65 (s), 32.15 ppm (s); 31P NMR (C6D6): d=
17.80 ppm (s); IR (CH2Cl2): n˜ = 1918 (CO) cmꢀ1; elemental analysis
calcd (%) for C34H32IrOP: C 60.07, H 4.74; found: C 60.14, H 4.55.
8a: 48%; 1H NMR (CDCl3): d=7.46 (d, J=7.8 Hz, 2H), 7.23–7.07 (m,
3H), 5.93 (d, J=7.5 Hz, 1H), 5.65 (t, J=7.5 Hz, 1H), 2.91 (d, J=7.5 Hz,
1H), 1.90 ppm (s, 3H); 13C NMR (CDCl3): d=145.91, 129.29, 129.15,
128.83, 128.25, 112.74, 111.60, 76.07, 28.38, 11.01 ppm; IR (Et2O): n˜ =
1865 (cyclopropene C=C) cmꢀ1
.
8b: 50%; 1H NMR (CDCl3): d=7.44 (d, J=7.7 Hz, 2H), 7.24–7.07 (m,
3H), 5.88 (d, J=7.5 Hz, 1H), 5.60 (t, J=7.5 Hz, 1H), 2.85 (d, J=7.5 Hz,
1H), 2.26 (q, J=7.5 Hz, 2H), 0.98 ppm (t, J=7.5 Hz, 3H); 13C NMR
(CDCl3): d=146.18, 129.42, 129.06, 128.86, 127.18, 117.90, 110.89, 75.82,
Thermolysis of 14d: Complex 14d (50 mg, 0.053 mmol) was dissolved in
benzene (3 mL) and heated at 508C for 15 h. Concentration of the so-
lution and chromatography of the residue as above afforded 15d (33 mg,
97%). The spectral data were identical to those given above.
28.17, 19.97, 12.26 ppm; IR (Et2O): n˜ = 1862 (cyclopropene C=C) cmꢀ1
.
Iridabenzene 14a: Cyclopropene 8a (250 mg, 0.89 mmol) was allowed to
react with BuLi (430 mL, 2.5m, 1.07 mmol) and [Ir(CO)Cl(PPh3)2]
(691 mg, 0.89 mmol) as described above for 13d. After warming to 08C
over 3 h, the reaction was filtered under an argon atmosphere to remove
the precipitated lithium salts. NMR analysis of the crude material
showed formation of a 3:2 mixture of 13a:14a along with protonated vi-
nylcyclopropene and a minute amount of 19a. Partial NMR data for 13a:
1H NMR (C6D6): d=6.25 (dq, J=8.2, 2.9 Hz, 1H), 3.10 (br s, 1H), 1.96
(s, 3H); 31P NMR (C6D6): d=0.43 (d, J=30.2 Hz), ꢀ1.23 ppm (d, J=
30.2 Hz). Partial NMR data for 19a: 1H NMR (C6D6): d=11.9 (q, J=
11.8 Hz, 1H), 8.29–8.23 ppm (m, 1H); 31P NMR (C6D6): d=21.17 ppm
(s).
8c: 58%; 1H NMR (CDCl3): d=7.51 (d, J=7.9 Hz, 2H), 7.45–7.29 (m,
3H), 5.99 (d, J=7.3 Hz, 1H), 5.82 (t, J=7.3 Hz, 1H), 3.00 (sept, J=
7.0 Hz, 1H), 2.65 (d, J=7.3 Hz, 1H), 1.33 ppm (d, J=7.0 Hz, 6H);
13C NMR (CDCl3): d=146.18, 129.17, 128.81, 128.60, 128.17, 121.63,
109.33, 75.29, 27.53, 26.94, 21.04, 20.89 ppm; IR (Et2O): n˜ = 1852 (cyclo-
propene C=C) cmꢀ1
.
8d: 72%; 1H NMR (CDCl3): d=7.51 (d, J=8.0 Hz, 2H), 7.46–7.31 (m,
3H), 6.00 (d, J=7.3 Hz, 1H), 5.85 (t, J=7.3 Hz, 1H), 2.67 (d, J=7.3 Hz,
1H), 1.35 ppm (s, 9H); 13C NMR (CDCl3): d=146.23, 129.32, 128.62
(2C), 128.16, 124.30, 107.88, 75.26, 32.14, 28.56, 27.39 ppm; IR (Et2O):
n˜ = 1854 (cyclopropene C=C) cmꢀ1
.
The filtrate from above was kept at 208C for 8 h and then cooled to
ꢀ308C overnight. The solid was collected and washed with cold Et2O to
give 14a as red crystals. Chromatography of the concentrated mother
liquor over neutral alumina (hexanes/Et2O, 3:1) gave additional material
(186 mg total, 23%). 14a: 1H NMR (C6D6): d=10.62 (q, J=10.8 Hz,
1H), 8.30 (q, J=6.3 Hz, 1H), 7.75 (dd, J=10.0, 7.9 Hz, 1H), 7.26–7.12
(m, 16H), 6.98–6.88 (m, 19H), 2.69 ppm (s, 3H); 13C NMR (C6D6): d=
204.97 (t, J=50.9 Hz), 187.46 (t, J=5.6 Hz), 183.92 (s), 170.02 (t, J=
3.5 Hz), 140.47 (t, J=8.1 Hz), 137.11–136.33 (m), 134.38 (t, J=5.5 Hz),
133.93 (t, J=5.5 Hz), 129.50 (s), 129.06 (t, J=3.8 Hz), 127.88 (s), 127.13
(s), 123.39 (s), 122.34 (t, J=7.1 Hz), 24.17 ppm (t, J=3.0 Hz); 31P NMR
(C6D6): d=18.03 ppm (s); IR (CH2Cl2): n˜ = 1982 (CO) cmꢀ1; elemental
analysis calcd (%) for C49H41IrOP2: C 65.39, H 4.59; found: C 64.82, H
4.59.
Iridabenzvalene 13d: A flamed-dried Schlenk flask was charged with cy-
clopropene 8d (240 mg, 0.74 mmol) and Et2O (20 mL) and cooled to
ꢀ788C. To this was added BuLi (330 mL, 2.5m, 0.82 mmol) at ꢀ788C
over 5 min. After stirring for 15 min at ꢀ788C, the resulting mixture was
transferred over 5 min by a double-ended needle to a ꢀ788C stirred sus-
pension of [Ir(CO)Cl(PPh3)2] (580 mg, 0.74 mmol) in Et2O (5 mL), and
then slowly warmed over a 3 h period to 08C. The solvent was removed
under vacuum and the residue purified by flash chromatography over
silica gel (hexanes/Et2O, 4:1), giving 13d (211 mg, 32%) as a pale yellow
solid. X-ray quality crystals of 13d were obtained by recrystallization
from a mixture of toluene and hexane. 1H NMR (C6D6): d=7.75–7.69
(dt, J=8.2, 1.5 Hz 6H), 7.50–7.41 (m, 6H), 7.36 (d, J=7.0 Hz, 2H), 7.08–
6.91 (m, 21H), 6.76–6.69 (m, 1H), 6.05 (dq, J=8.1, 2.7 Hz, 1H), 3.20 (br
s, 1H), 1.11 ppm (s, 9H); 13C NMR (C6D6): d=181.86 (dd, J=7.9,
1198
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 1191 – 1200

Metallabenzenes and Valence Isomers
FULL PAPER
Iridabenzene 14b: Cyclopropene 8b (200 mg, 0.68 mmol) was allowed
toreact with BuLi (330 mL, 2.5m, 0.81 mmol) and [Ir(CO)Cl(PPh3)2]
(530 mg, 0.68 mmol) as described above for 14a. NMR analysis of the
crude material showed formation of a 10:3:1 mixture of 13b:14b:19b
along with protonated vinylcyclopropene. Partial NMR data for 13b:
1H NMR (C6D6): d=6.25 (dq, J=8.0, 2.8 Hz, 1H), 3.20 (br s, 1H);
31P NMR (C6D6): d=ꢀ0.05 (d, J=31.8 Hz), ꢀ1.38 ppm (d, J=31.8 Hz).
Partial NMR data for 19b: 1H NMR (C6D6): d=11.20 (q, J=11.7 Hz,
1H), 8.18 (q, J=6.3 Hz, 1H), 1.49 ppm (t, J=7.3 Hz, 3H); 31P NMR
(C6D6): d=20.61 ppm (s).
15.1 Hz), 127.17 (s), 127.14 (s), 123.40 (s), 71.03 (dd, J=72.5, 4.0 Hz),
65.88 (dd, J=70.5, 3.0 Hz), 58.80 (t, J=4.0 Hz), 28.18 (t, J=4.0 Hz),
20.85 (dd, J=30.2 Hz, 3.0 Hz), 20.07 (dd, J=30.2, 2.0 Hz), 14.71 ppm (d,
J=3.0 Hz); 31P NMR (C6D6): d=ꢀ52.11 (d, J=32.7 Hz), ꢀ54.62 ppm (d,
J=32.7 Hz); IR (CH2Cl2): n˜ = 1966 (CO) cmꢀ1; elemental analysis calcd
(%) for C20H31IrOP2: C 44.35, H 5.77; found: C 43.89, H 5.79.
Iridabenzvalene 20d: Reaction of cyclopropene 8d (240 mg, 0.74 mmol)
with BuLi (0.33 mL, 2.5m, 0.82 mmol) and [Ir(CO)Cl(PMe3)2] (302 mg,
0.74 mmol) as described for 13d gave benzvalene 20d (244 mg, 58%) as
light yellow crystals. 20d: 1H NMR (C6D6): d=7.63 (d, J=7.3 Hz, 2H),
7.20 (t, J=7.6 Hz, 2H), 7.13–7.06 (m, 1H), 7.01 (t, J=7.0 Hz, 1H), 6.43–
6.37 (m, 1H), 3.32 (br s, 1H), 1.34 (s, 9H), 1.29 (d, J=9.0 Hz, 9H),
1.04 ppm (d, J=9.0 Hz, 9H); 13C NMR (C6D6): d=181.02 (t, J=7.1 Hz),
147.99 (m), 146.34 (d, J=8.1 Hz), 140.00 (t, J=16.1 Hz), 127.79 (s),
127.76 (s), 123.46 (s), 83.14 (dd, J=73.5, 4.0 Hz), 65.83 (dd, J=66.5,
2.5 Hz), 54.87 (t, J=4.0 Hz), 35.37 (t, J=4.0 Hz), 31.24 (d, J=2.0 Hz),
21.70 (dd, J=30.2, 4.0 Hz), 19.35 ppm (dd, J=29.2, 2.0 Hz); 31P NMR
(C6D6): d=ꢀ53.98 (d, J=35.1 Hz), ꢀ55.38 ppm (d, J=35.1 Hz); IR
The filtrate of the above mixture was kept at 208C for 8 h and then
cooled to ꢀ308C overnight. The solid was collected and washed with
cold Et2O to give 14b as red crystals. Careful chromatography of the
mother liquor over neutral alumina (hexanes/Et2O, 3:1) furnished addi-
tional pure 14b (131 mg total, 21%). 14b: 1H NMR (C6D6): d=10.61 (q,
J=10.9 Hz, 1H), 8.29 (q, J=6.1 Hz, 1H), 7.79 (dd, J=10.2, 7.8 Hz, 1H),
7.32 (d, J=7.0 Hz, 2H), 7.23–7.12 (m, 14H), 6.96–6.87 (m, 19H), 3.01 (q,
J=7.4 Hz, 2H), 1.35 ppm (t, J=7.4 Hz, 3H); 13C NMR (C6D6): d=204.17
(t, J=50.4 Hz), 187.31 (t, J=5.5 Hz), 184.95 (s), 169.35 (t, J=3.5 Hz),
140.84 (t, J=6.0 Hz), 139.99 (t, J=8.1 Hz), 137.14–136.36 (m), 134.38 (t,
J=5.5 Hz), 129.48 (s), 129.36 (t, J=4.6 Hz), 127.89 (s), 126.66 (s), 123.34
(s), 122.90 (t, J=6.5 Hz), 29.52 (t, J=3.1 Hz), 19.23 ppm (s); 31P NMR
(C6D6): d=18.65 ppm (s); IR (CH2Cl2): n˜ = 1892 (CO) cmꢀ1; elemental
analysis calcd (%) for C50H43IrOP2: C 65.70, H 4.74; found: C 65.42, H
4.92.
(CH2Cl2): n˜ = 1960 (CO) cmꢀ1
; elemental analysis calcd (%) for
C22H35IrOP2: C 46.38, H 6.19; found: C 45.96, H 6.19.
Iridabenzene 21b: A solution of 20b (54 mg, 0.1 mmol) in degassed C6D6
(1 mL) was placed into an NMR tube and heated at 758C over 24 h, re-
sulting in complete isomerization. The solvent was removed and the solid
redissolved in 1:1 hexane/Et2O. After the mixture had been cooled at
ꢀ308C overnight, the red solid was collected and washed with cold Et2O
to give 21b (48 mg, 89%) as air-sensitive red crystals. 21b: 1H NMR
(C6D6): d=10.92 (dq, J=11.4. 1.2 Hz, 1H), 8.13 (dq, J=6.2, 1.2 Hz, 1H),
7.90 (dd, J=9.6. 8.4 Hz, 1H), 7.43–7.31 (m, 4H), 7.08 (t, J=7.1 Hz, 1H),
2.85 (q, J=7.5 Hz, 1H), 1.26 (t, J=7.5 Hz, 3H), 1.07 ppm (d, J=10.3 Hz,
18H); 13C NMR (C6D6): d=190.79 (t, J=50.4 Hz), 188.84 (t, J=5.0 Hz),
174.20 (s), 169.29 (t, J=3.0 Hz), 138.56 (t, J=5.0 Hz), 135.38 (t, J=
8.1 Hz), 129.70 (t, J=4.0 Hz), 126.23 (s), 123.86 (t, J=6.0 Hz), 123.08 (s),
29.32 (t, J=4.0 Hz), 20.91 (dd, J=14.3, 2.0 Hz), 18.82 ppm (s); 31P NMR
(C6D6): d=ꢀ38.56 ppm (s); elemental analysis calcd (%) for
C20H31IrOP2: C 44.35, H 5.77; found: C 42.98, H 5.67.
Iridabenzene 14c: Cyclopropene 8c (141 mg, 0.50 mmol) was allowed to
react with BuLi (240 mL, 2.5m, 0.60 mmol) and [Ir(CO)Cl(PPh3)2]
(390 mg, 0.50 mmol) as described above for 14a. NMR analysis of the
crude material showed formation of a 13:2:1 mixture of 13c:14c:19c
along with protonated vinylcyclopropene. Attempts to isolate/purify 13c
resulted in isomerization to 14c. Partial NMR data for 13c: 1H NMR
(C6D6): d=6.22 (dq, J=8.2, 2.9 Hz, 1H), 2.80 ppm (br s, 1H); 31P NMR
(C6D6): d=ꢀ0.86 (d, J=30.5 Hz), ꢀ2.03 ppm (d, J=30.5 Hz). Partial
NMR data for 19c: 1H NMR (C6D6): d=10.75 ppm (q, J=11.8 Hz, 1H);
31P NMR (C6D6): d=20.11 ppm (s).
The filtrate of the above mixture was kept at 208C for 30 h and then
cooled to ꢀ308C overnight. The solid was collected and washed with
cold Et2O to give 14c (134 mg, 29%) as red crystals. 14c: 1H NMR
(C6D6): d=10.61 (q, J=10.9 Hz, 1H), 8.41 (q, J=6.3 Hz, 1H), 7.85 (dd,
J=10.0, 7.9 Hz, 1H), 7.29 (d, J=7.7 Hz, 2H), 7.23–7.11 (m, 14H), 6.97–
6.88 (m, 19H), 3.56 (sept, J=6.7 Hz, 1H), 1.41 ppm (d, J=6.7 Hz, 6H);
13C NMR (C6D6): d=203.52 (t, J=50.4 Hz), 187.27 (t, J=6.0 Hz), 185.14
(s), 169.68 (t, J=3.2 Hz), 145.14 (t, J=6.0 Hz), 137.01–136.93 (m), 136.57
(t, J=8.0 Hz), 134.40 (t, J=5.5 Hz),
Iridabenzene 21d and cyclopentadienyliridium complex 22d: A solution
of 20d (57 mg, 0.1 mmol) in degassed C6D6 (1 mL) was heated at 758C
for 48 h leading to a mixture of unreacted 20d, 21d, and 22d (5:10:1)
along with numerous other decomposition products. Continue heating
led to complete decomposition of the sample. 21d: 1H NMR (C6D6): d=
10.87 (q, J=12.0 Hz, 1H), 8.46 (q, J=6.2 Hz, 1H), 7.88 (d, J=9.7 7.9 Hz,
1H), 7.66–6.94 (m, 5H), 1.47 (s, 9H), 1.05 ppm (d, J=10.0 Hz); 31P NMR
129.47 (s), 128.92 (s), 126.75 (s), 123.28
(s), 122.58 (t, J=7.0 Hz), 30.44 (t, J=
Table 4. Crystal data for compounds 13d, 14a, 14b, and 14d.
3.0 Hz), 26.54 ppm (s) (1 aromatic res-
onance obscured); 31P NMR (C6D6):
d=18.79 ppm (s); IR (CH2Cl2): n˜ =
13d
14a
14b
14d
mol formula
mol. wt.
crystal system
space group
a [ꢂ]
b [ꢂ]
c [ꢂ]
a [8]
b [8]
C52H47IrP2·2C7H8
1126.4
monoclinic
I2/a
22.602(3)
11.3221(13)
23.853(5)
90
106.01(2)
90
5867(2)
4
C49H41IrOP2
900.0
C50H43IrOP2
914.1
C52H47IrOP2
942.1
1886 (CO) cmꢀ1
; elemental analysis
triclinic
triclinic
P1
monoclinic
P21/c
calcd (%) for C51H45IrOP2: C 66.00, H
4.89; found: C 65.75, H 4.86.
¯
¯
P1
10.979(3)
10.983(3)
18.209(3)
100.69(2)
95.74(2)
110.79(3)
1983.9(11)
2
10.880(2)
11.130(2)
17.883(2)
89.59(1)
83.42(1)
70.65(2)
2028.6(8)
2
11.626(2)
20.564(5)
18.529(5)
90
106.75(2)
90
Iridabenzvalene 20b: Reaction of cy-
clopropene 8b (220 mg, 0.74 mmol)
with BuLi (0.33 mL, 2.5m, 0.82 mmol)
and
[Ir(CO)Cl(PMe3)2]
(302 mg,
0.74 mmol) as described for 13d gave
benzvalene 20b (164 mg, 41%) as
light yellow crystals. 20b: 1H NMR
(C6D6): d=7.64 (d, J=8.5 Hz, 2H),
7.28 (t, J=7.9 Hz, 2H), 7.21–7.14 (m,
1H), 7.06 (t, J=7.0 Hz, 1H), 6.50–6.40
(m, 1H), 3.34 (br s, 1H), 2.61 (t, J=
7.3 Hz, 2H), 1.24 (d, J=9.0 Hz, 9H),
1.20 (t, J=7.3 Hz, 3H), 1.10 ppm (d,
J=9.0 Hz, 9H); 13C NMR (C6D6): d=
179.72 (t, J=7.6 Hz), 147.67 (m),
146.40 (t, J=4.0 Hz), 140.17 (t, J=
g [8]
V [ꢂ3]
4242.2(2)
4
Z
F000
2296
900
916
1896
2qmax [8]
46
23
48
23
independent reflections
observed [Iꢂs(I)]
used in refinement
refined parameters
R(F)/wR [Iꢂs(I)]
R(F2)/wR(F2) (all)
4386
4095
3591
245
0.063, 0.074
0.112, 0.146
6233
6229
5160
478
0.050, 0.047
0.087, 0.098
6354
7688
6352
7441
5280
4835
487
505
0.036, 0.041
0.062, 0.086
0.063, 0.052
0.097, 0.103
Chem. Eur. J. 2005, 11, 1191 – 1200
www.chemeurj.org
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1199

M. M. Haley et al.
(C6D6): d=ꢀ38.79 ppm (s). Partial data for 22d: d=5.26 (br s, 1H), 4.72
[11] P. Mꢅller, C. Grꢆnicher, Helv. Chim. Acta 1993, 76, 521–534.
(br s, 1H), 4.54 (br s, 1H), 1.35 ppm (s, 9H).
[12] P. Mꢅller, C. Grꢆnicher, Helv. Chim. Acta 1995, 78, 129–144.
[13] W. A. Donaldson, R. P. Hughes, J. Am. Chem. Soc. 1982, 104, 4846–
4859.
[14] Insertion of carbene into benzene: a) G. Maas, Top Curr. Chem.
1987, 137, 75–253; b) A. J. Anciaux, A. Demonceau, A. J. Hubert,
A. F. Noels, N. Petiniot, P. Teyssie, J. Chem. Soc. Chem. Commun.
1980, 765–766.
[15] G. Maier, M. Hoppe, H. P. Reisenauer, C. Kruger, Angew. Chem.
1982, 94, 445; Angew. Chem. Int. Ed. Engl. 1982, 21, 437.
[16] D. B. Dess, J. C. Martin, J. Am. Chem. Soc. 1991, 113, 7277–7287.
[17] a) G. Stork, K. Zhao, Tetrahedron Lett. 1989, 30, 2173–2174; b) H. J.
Bestmann, H. G. Rippel, R. Dostalek, Tetrahedron Lett. 1989, 30,
5261–5262.
[18] a) C. E. Housecroft, A. G. Sharpe, Inorganic Chemistry, Prentice
Hall, London, 2001; b) G. O. Spessard, G. L. Miessler, Organometal-
lic Chemistry, Prentice Hall, London, 1997; c) J. P. Collman, L. S.
Hegedus, J. R. Norton, R. G. Finke, Principles and Applications of
Organotransition Metal Chemistry, University Science Books, Mill
Valley, CA, 1987.
Kinetics of the rearrangement of iridabenzene 14d to cyclopentadienyl-
iridium complex 15d: Solutions of 14d in C6D6 were prepared in three
NMR tubes under N2 and heated at 40, 50, and 608C, respectively. The
1
reaction progress was monitored by H NMR spectroscopy. Integration of
the proton resonance signals was used to calculated ln[C]/[C]o. A plot of
ln[C]/[C]o versus time gave a straight line for the data at each different
temperature. The first order constants were obtained from the slopes of
these lines.
X-ray structure determinations: Data were collected on an Enraf-Nonius
CAD-4 Turbo diffractometer using MoKa radiation, l=0.71073 ꢂ; graph-
ite monochromator; T=296 K; scan mode w–2q. Pertinent crystallo-
graphic data and refinement parameters are given in Table 4. Structure
refinement (C atoms anisotropic, H atom riding) was accomplished with
the teXsan program suite (version 1.7 for SGI workstations). CCDC-
218966 (13d), CCDC-218968 (14a), CCDC-218967 (14b), and CCDC-
218965 (14d) contain the supplementary crystallographic data for this
paper. These data can be obtained free of charge via www.ccdc.cam.
ac.uk/conts/retrieving.html (or from the Cambridge Crystallographic
Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+44)
1223-336-033; or deposit@ccdc.cam.ac.uk).
[19] a) H. M. McConnell, J. Chem. Phys. 1957, 27, 226; b) A. J. Ashe III,
Acc. Chem. Res. 1978, 11, 153–157; c) P. Jutzi, Angew. Chem. 1975,
87, 269–283; Angew. Chem. Int. Ed. Engl. 1975, 14, 232–245; d) P.
Schwab, R. H. Grubbs, J. W. Ziller, J. Am. Chem. Soc. 1996, 118,
100–110.
[20] Berry-type process for isomerization of PR3 on metal: F. A. Cotton,
G. Wilkinson, Advanced Inorganic Chemistry, 4th ed., Wiley, New
York, 1980, pp. 1218–1220.
[21] a) R. V. Williams, L. D. Shirtcliff, D. A. Braden, R. Herges, M. M.
Haley, unpublished results. Such calculations do not account for sol-
vent effects; b) see also: M. A. Iron, A. C. B. Lucassen, H. Cohen,
M. E. van der Boom, J. M. L. Martin, J. Am. Chem. Soc. 2004, 126,
11699–11710.
Acknowledgement
This work was supported by the National Science Foundation (CHE-
0075246). We thank Dr. Alex Blumenfeld (University of Idaho) for ac-
quisition of the HMBC and HMQC NMR spectra and Dima Azar for
her help in synthesizing 9c–12c.
[22] Inter alia: a) R. Ferede, N. T. Allison, Organometallics 1983, 2, 463–
465; b) R. R. Schrock, S. F. Pederson, M. R. Churchill, J. W. Ziller,
Organometallics 1984, 3, 1574–1583; c) R. Ferede, J. F. Hinton,
W. A. Korfmacher, J. P. Freeman, N. T. Allison, Organometallics
1985, 4, 614–616; d) C. A. Mike, R. Ferede, N. T. Allison, Organo-
metallics 1988, 7, 1457–1459; e) J. Yang, J. Yin, K. A. Abbound,
W. M. Jones, Organometallics 1994, 13, 971–978; f) R. P. Hughes,
A. S. Kowalski, Organometallics 1998, 17, 270–273.
[1] J. R. Bleeke, Chem. Rev. 2001, 101, 1205–1227.
[2] a) G. P. Elliott, W. R. Roper, J. M. Waters, J. Chem. Soc. Chem.
Commun. 1982, 811–813; b) G. P. Elliott, N. M. McAuley, W. R.
Roper, Inorg. Synth. 1989, 26, 184–189.
[3] J. R. Bleeke, R. Behm, Y.-F. Xie, M. Y. Chiang, K. D. Robinson,
A. M. Beatty, Organometallics 1997, 16, 606–623, and references
therein.
[4] H. Xia, G. He, H. Zhang, T. B. Wen, H. H. Y. Sung, I. D. Williams,
G. Jia, J. Am. Chem. Soc. 2004, 126, 6862–6863.
[23] J. Yang, W. M. Jones, J. K. Dixon, N. T. Allison, J. Am. Chem. Soc.
1995, 117, 9776–9777.
[5] M. Paneque, C. M. Posadas, M. L. Poveda, N. Rendꢄn, V. Salazar, E.
OÇate, and K. Mereiter, J. Am. Chem. Soc. 2003, 125, 9898–9899.
[6] a) R. D. Gilbertson, T. J. R. Weakley, M. M. Haley, J. Am. Chem.
Soc. 1999, 121, 2597–2598; b) R. D. Gilbertson, T. L. S. Lau, S.
Lanza, H.-P. Wu, T. J. R. Weakley, M. M. Haley, Organometallics
2003, 22, 3279–3289.
[7] R. D. Gilbertson, T. J. R. Weakley, M. M. Haley, Chem. Eur. J. 2000,
6, 437–441.
[8] H.-P. Wu, T. J. R. Weakley, M. M. Haley, Organometallics 2002, 21,
4320–4322.
[9] a) V. Jacob, T. J. R. Weakley, M. M. Haley, Angew. Chem. 2002, 114,
3620–3623; Angew. Chem. Int. Ed. 2002, 41, 3470–3473; b) C. W.
Landorf, V. Jacob, T. J. R. Weakley, M. M. Haley, Organometallics
2004, 23, 1174–1176; c) see also: V. Jacob, T. J. R. Weakley, M. M.
Haley, Organometallics 2002, 21, 5394–5400.
[24] M. A. Iron, J. M. L. Martin, M. E. van der Boom, J. Am. Chem. Soc.
2003, 125, 13020–13021.
[25] Relevant examples, for example: a) R. S. Threlkel, J. E. Bercaw, J.
Am. Chem. Soc. 1981, 103, 2650–2659; b) J. F. Hoover, J. M. Stryker,
J. Am. Chem. Soc. 1990, 112, 464–465; c) R. L. Trace, J. Sanchez, J.
Yang, J. Yin, W. M. Jones, Organometallics 1992, 11, 1440–1442;
d) G. L. Casty, J. M. Stryker, Organometallics 1997, 16, 3083–3085.
[26] N. M. Libman, O. N. Shandybina, J. Org. Chem. USSR 1984, 20,
2290–2294.
[27] G. Capozzi, R. Ottana, G. Romeo, F. Marcuzzi, Gazz. Chim. Ital.
1985, 115, 311–314.
[28] M. J. Burk, R. H. Crabtree, Inorg. Chem. 1986, 25, 931–932.
[10] H.-P. Wu, S. Lanza, T. J. R. Weakley, M. M. Haley, Organometallics
2002, 21, 2824–2826.
Received: August 31, 2004
Published online: December 29, 2004
1200
ꢁ 2005 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2005, 11, 1191 – 1200