Aminoethyl-functionalized cyclopentadienyliridium complexes: Photochemical C-H activation and carbonylation of cycloalkanes

Article abstract of DOI:10.1002/ejic.200501049

The photochemical C-H activation of cycloalkanes by aminoethyl-functionalized cyclopentadienyliridium complexes proceeds in a broadly similar way to that for their unfunctionalized analogues. In a carbon monoxide atmosphere, there is evidence that carbonylation to form the carboxaldehyde occurred with both systems. In situ infrared spectroscopic studies on these reactions allow the identification of several species present in low concentrations in the reaction mixture. Wiley-VCH Verlag GmbH & Co. KGaA, 2006.

Full text of DOI:10.1002/ejic.200501049

FULL PAPER
DOI: 10.1002/ejic.200501049
Aminoethyl-Functionalized Cyclopentadienyliridium Complexes: Photochemical
C–H Activation and Carbonylation of Cycloalkanes
Pek Ke Chan,[a] Weng Kee Leong,*[a] Karl I. Krummel,[b] and Marc V. Garland*[b]
Keywords: Carbonylation / C–H activation / Cyclopentadienyl ligands / Iridium / Photochemistry
The photochemical C–H activation of cycloalkanes by amino-
occurred with both systems. In situ infrared spectroscopic
ethyl-functionalized cyclopentadienyliridium complexes pro- studies on these reactions allow the identification of several
ceeds in a broadly similar way to that for their unfunction-
alized analogues. In a carbon monoxide atmosphere, there
is evidence that carbonylation to form the carboxaldehyde
species present in low concentrations in the reaction mixture.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2006)
be strengthened by the chelate effect.[4] As a result, amino-
ethyl-functionalized cyclopentadienyl ligands may be ex-
pected to behave as hemilabile ligands; under the conditions
for the activation of hydrocarbons, the nitrogen donor on
the side-arm of the aminoethyl-functionalized cyclopen-
tadienyliridium complex may coordinate to the iridium cen-
ter. This reversible coordination to a reactive metal center
is of general interest in catalysis as it may stabilize a highly
reactive, electronically and sterically unsaturated intermedi-
ate by weakly occupying the vacant coordination site until
the actual substrate coordinates and replaces the amino
group.[5] In this paper, we present the results of our study
on the photolysis of [Cp*Ir(CO)2] (1b), as well as its amino-
ethyl-functionalized analogue, in vacuo or under an argon
or carbon monoxide atmosphere.
Introduction
UV initiated photodissociation of CO or H2 from group-
9 organometallic compounds of the type [CpMLLЈ] or
[Cp*MLLЈ] (M = Ir, Rh; L = CO, PMe3; LЈ = CO, H2)
have been known to generate intermediates that are capable
of activating the otherwise inert C–H bonds in hydrocarbon
solvents.[1] This has stimulated much interest due to the
plentiful supply of alkanes and the potential of using al-
kanes as chemical feedstocks for catalytic syntheses of or-
ganic molecules. Over the last two decades there have been
several studies on the mechanism, kinetics, and thermody-
namics of this activation process in solution, the gas phase,
low-temperature matrices, and liquid noble gases.[1d,2] For
[CpXM(CO)2] (CpX = Cp or Cp*; M = Ir or Rh), spectro-
scopic measurements have shown that the primary
photoproduct is a coordinatively unsaturated 16-electron
species [CpXM(CO)] obtained by dissociation of a CO
ligand. This species is extremely reactive and forms a
solvent adduct complex [CpXM(CO)···S] before undergoing
C–H activation to give the hydridoalkyl species
[CpXM(CO)(R)(H)].[2b,3]
Many functionalized cyclopentadienyl ligands containing
side-chain donor groups (OR, NRRЈ, PRRЈ, CH=CH2) are
beginning to appear in the literature. In particular, we were
interested in aminoethyl-functionalized cyclopentadienyl li-
gands because of their “hard-soft” combination of electron
donors; the cyclopentadienyl ligand is known to stabilize
transition metals in high as well as low oxidation states
whereas the amino group favors coordination to metals in
high oxidation states. Only weak interactions are antici-
pated to metals in low oxidation states, although these may
Results and Discussion
The aminoethyl-functionalized cyclopentadienyliridium
∧
∧
∧
complexes [Cp Ir(CO)2] (1c) and [Cp* Ir(CO)2] (1d; Cp
∧
= C5H4CH2CH2NMe2 and Cp* = C5Me4CH2CH2NMe2)
were synthesized by an analogous route to that reported for
[CpIr(CO)2] (1a; Scheme 1).[6]
The UV absorption spectrum of 1d shows absorption
maxima at about 210 and 280 nm. Upon UV irradiation
of a degassed solution of 1d in cyclohexane, the solution
darkened slowly from yellow to orange-brown. Monitoring
the photolysis by infrared spectroscopy showed initial for-
mation only of the corresponding hydridoalkyl species 2d
(νCO = 1981 cm–1), the CO stretching vibration of which is
essentially identical to that of the [Cp*IrCO(Cy)(H)] ana-
logue (νCO = 1981 cm–1 in cyclohexane).[7] As the photolysis
progressed, a new species showing a carbonyl stretch at
1996 cm–1 in the infrared spectrum and a hydride resonance
[a] Department of Chemistry, National University of Singapore,
3 Science Drive 3, Singapore 117543, Singapore
[b] Department of Chemical and Biomolecular Engineering,
National University of Singapore,
1
at δ = –15.7 ppm ([D8]toluene) in the H NMR spectrum
4 Engineering Drive 4, Singapore 119260, Singapore
was formed (Scheme 2). This was assigned to the dihydride
1568
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1568–1572

Aminoethyl-Functionalized Cyclopentadienyliridium Complexes
FULL PAPER
group on the Cp ligand or CO insertion into the Ir–C bond
∧
to give the acyl species [Cp* Ir(CO)(H)(COC6H11)].
Under a CO atmosphere (1 bar), the activation of 1d in
cyclohexane to form 2d was slower and occurred with lower
yield; the ratio of the absorbance of the CO stretch for 2d/
1d was 0.43 after 5 h under a CO atmosphere, and 1.4 after
2 h in vacuo. The same difference in ratios was observed for
the activation of 1b in cyclohexane to form 2b; the ratio of
(absorbance of) 2b/1b was essentially constant at 0.65 after
30 min under a CO atmosphere, while it increased steadily
from 1.03 at 30 min to 1.61 at 60 min in vacuo, although
there were signs of decomposition in the latter on pro-
longed photolysis. Instead of 3, the cluster [Ir4(CO)12] (4;
νCO = 2029, 2069 cm–1) was obtained together with 2d; clus-
ter 4 was also obtained from 1b under similar conditions.
Cluster 4 was not formed when a cyclohexane solution of
1d was stirred under a CO atmosphere without irradiation.
However, if the cyclohexane solution of 1d was first irradi-
ated, and then a CO atmosphere introduced, a mixture of
1d, 2d, and 4 was obtained; the changes in the relative inten-
sities in the infrared spectra suggest that 3 is completely
converted into 4 while 2d is only partially converted into 4.
If this mixture was degassed and then subjected to irradia-
tion, a mixture of 1d, 2d, and 3 was obtained. All these are
summarized by the scheme and spectra given in Figure 1,
and suggest that 4 is formed by the decomposition of 2d
and 3, and that this is reversible.
Scheme 1.
complex 3 by comparison of its spectroscopic data with the
analogue [Cp*Ir(CO)H2] (νCO = 1996 cm–1 in cyclohexane;
δ = –15.8 ppm in C6D6);[7,8] the latter was reported to form
when 1b was irradiated under a hydrogen atmosphere.[7]
Scheme 2.
Since our irradiation experiment was carried out in the
absence of hydrogen gas, the formation of 3 may have re-
sulted from β-hydrogen elimination of cyclohexene from 2d.
Indeed, cyclohexene was detected by GC-MS analysis of
the reaction solution. Similar irradiation of 1c in cyclohex-
ane also resulted in the formation of the corresponding di- Figure 1. IR spectra of 1d taken in cyclohexane: (a) after UV irradi-
ation in vacuo for 6 h, followed by (b) stirring overnight under 1
hydride species, although in this case there appeared to be
atm of CO, (c) degassing and stirring overnight in vacuo, and (d)
UV irradiation in vacuo for 3 h.
some decomposition during irradiation. Similar results
could also be achieved in a shorter time by irradiation with
a 266-nm laser; for 1c, the ratio of 1c/2c, as determined by
We confirmed the identity of 4 by an independent syn-
the relative intensities of the CO absorbance, was 3.3 after thesis,[9] and also found that upon irradiation, a degassed
6 h of UV irradiation, while the ratio was 2.0 after irradia- suspension of 4 in cyclohexane in the presence of Cp*H or
∧
tion with the laser for 45 minutes. We also found that with Cp* H gave partial conversion to 1b or 1d, respectively,
[(C5H4CH2Ph)Ir(CO)2] (1e), which has a benzyl group as as identified by IR spectroscopy. Further irradiation of the
the side-arm, a dihydride species was also obtained. It reaction mixture containing 1b and the starting materials
therefore appears that photolysis of the complexes carrying resulted in the formation of the corresponding hydridoalkyl
a side-arm results in the formation of dihydride complexes; species 2b; the yield of 2d for the corresponding reaction
this has not been reported to occur with [Cp*Ir(CO)2] itself. was low. At the end of the 5 h irradiation, suspension of
1
The H NMR spectrum of the reaction mixture also shows 4 in the reaction mixture was still observed, thus the low
another hydride resonance at δ = –17.5 ppm, which could conversion rates may be attributed to the low solubility of
be due to intramolecular C–H activation of the amine 4 in cyclohexane.
Eur. J. Inorg. Chem. 2006, 1568–1572
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1569

P. K. Chan, W. K. Leong, K. I. Krummel, M. V. Garland
FULL PAPER
In line with earlier findings that smaller rings show ful for picking up signals due to species that are present in
higher reactivity toward C–H activation,[1c,10] we found that low concentration, have low absorbance, are unstable, or
the C–H activation of cyclopentane by 1d was indeed more have peaks that overlap with those due to other species.[11]
facile; the ratio of 2e/1d (by absorbance) was 2.22 after The schematic diagram for the set-up used for the in situ
10 min, compared to a 2d/1d ratio of 1.94 after the same IR measurements is shown below (Figure 3).
period. These figures also indicate that the activation of 1d
Measurements were made for both 1b and 1d; for each
is faster than that of 1b. On stirring the resultant solution complex, spectra were collected at 5 min intervals, and a
overnight under a CO atmosphere, both 2e and 3 converted total of 232 spectra were collected.[12] For 1b, the individual
completely into 4 (Figure 2).
IR spectra of 1b, 2b, and 4, in the CO stretching region
(1650–2250 cm–1) were recovered, together with the C=O
stretch for cyclohexanecarboxaldehyde (Figure 4). Further
confirmation of the aldehyde identity was obtained by
NMR and GC-MS characterization. Formation of the cor-
responding carboxaldehyde in reactions carried out with cy-
clopentane was also observed by NMR spectroscopy. The
carboxaldehyde could have been formed by carbonylation
Figure 2. IR spectrum of 1d in cyclopentane: (a) after stirring for
2 h under CO, followed by (b) 3 h UV irradiation under CO, (c)
degassing and irradiation for 3 h in vacuo, and finally (d) stirring
overnight under 1 atm of CO.
To gain further insight into the process, particularly with
respect to the detection of compounds that we have hitherto
not isolated or identified, we carried out in situ IR measure-
ments on the C–H activation of cyclohexane, and used the
band-targeted entropy minimization (BTEM) technique
that one of us has developed for deconvolution of the data
Figure 4. Pure component IR spectra of individual species
recovered from deconvolution of the IR spectra of the reaction
mixture: (1) [Cp*Ir(CO)2] (1b), (2) [Cp*Ir(CO)(H)(C6H11)] (2b), (3)
matrix.[11] The BTEM algorithm has been proven to be use- [Ir4(CO)12] (4), and (4) C6H11CHO.
Figure 3. Schematic diagram of the set-up used for in situ IR measurements.
1570
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1568–1572

Aminoethyl-Functionalized Cyclopentadienyliridium Complexes
FULL PAPER
with an HP 5973 mass-selective detector and
a
ZB-1
of the hydridoalkyl species followed by reductive elimi-
nation. Carbonylation of hydrocarbons catalyzed by group-
9 transition metal complexes such as [RhCl(CO)(PMe3)3]
under UV irradiation has been reported,[13] but 18-electron
Cp*Ir complexes have not been known to carbonylate hy-
drocarbons as it was believed that coordination of an in-
coming CO ligand would result in dissociation of the hydro-
carbon.[14] A FAB-MS analysis of the reaction mixture
yielded a fragment at m/z 267 in the positive-ion mode, and
the 13C NMR spectrum also shows a resonance at δ =
210.54 ppm, both of which suggest the presence of an acyl
species, which may be the intermediate in the carbonylation
reaction, although we have yet to confirm its identity.
We analyzed the yield of carboxaldehyde by GC-FID
and found that UV irradiation of 1b in cyclohexane in
vacuo for one hour (absence of external CO source) af-
forded 0.9% of cyclohexanecarboxaldehyde (wrt [Ir]). When
the photolysis was carried out under CO (1 bar) for the
same length of time, the yield of aldehyde was 71.8%. So
far, however, the yield is still far from catalytic. A high CO
pressure may inhibit the initial CO dissociation from the
dicarbonyliridium complex while favoring the insertion of
CO into the Ir–C bond to form the putative [Cp*Ir(CO)-
(COC6H11)(H)] acyl intermediate. Photolysis at various CO
pressures (1.3 to 7.4 bars) was carried out but we did not
find any significant trend in the aldehyde yield with CO
pressure.
(30 m×0.25 mm×0.25 µm) capillary column, or with an HP5890
Series II plus equipped with an FID detector and a DB-5
(30 m×0.32 mm×0.25 µm) capillary column. Photolyses were car-
ried out with either a Hanovia 450-W UV lamp with a nominal
λmax of 254 nm, or a continuum Surelite III 10-ns-pulse Nd-YAG
laser operating at 3 mJ per pulse with 266-nm radiation.
IR spectra for routine analysis were recorded with a Bio-Rad FTS
165, a Digilab Excalibur Series FTS 3000MX, or a Shimadzu IR
Prestige-21 FTIR-8400S FT-IR spectrometer at a resolution of 1
cm–1 using a solution IR cell with NaCl windows and a path length
of 0.1 mm. IR spectra for in situ studies under ambient pressure
were recorded with a Perkin–Elmer 2000 FTIR spectrometer at a
resolution of 4 cm–1 using a high-pressure heatable liquid cell with
ZnSe windows and a path length of 0.1 mm from Specac. IR spec-
tra for high-pressure studies were taken in a thermostatted high-
pressure cell constructed of 316 stainless steel and AMTIR win-
dows with a path length of 0.5 mm; details of the design will be
described elsewhere.[15] Only carbonyl stretches in the 1600–
2200 cm–1 region are reported.
∧
∧
The compounds Cp H,[16] Cp* H,[17] C5H5CH2Ph (CpBzH),[18]
and [Cp*Ir(CO)2] (1b)[19] were prepared following published pro-
cedures. Thallium salts were prepared following
a literature
∧
∧
method.[20] The complexes [Cp Ir(CO)2] (1c), [Cp* Ir(CO)2] (1d),
and [CpBzIr(CO)2] (1e) were synthesized by a procedure analogous
to that used for the synthesis of [CpIr(CO)2] (1a).[6]
∧
∧
Preparation of [Cp* Ir(CO)2]: Cp* Li was prepared in situ from
∧
Cp* H (98.2 mg, 0.509 mmol) and nBuLi (0.3 mL of a 2.2  solu-
tion, 0.66 mmol) in hexane. The suspension was cannula transfer-
red into a suspension of [Ir(CO)3Cl] (102 mg, 0.327 mmol) in hex-
ane (10 mL) in a Carius tube. The mixture was degassed and heated
at 70 °C for 2 d. The resultant mixture was cooled and filtered
through celite to remove unreacted starting materials. The yellow
filtrate was vacuum-dried to give 1d as a yellow oil. IR (cyclohex-
Conclusions
We have found that dicarbonylcyclopentadienyliridium
complexes in which there is a side-arm carrying an amino
group undergo photochemical C–H activation of alkanes in
much the same way as the parent complexes without the
1
ane): ν = 1954, 2020 cm–1. H NMR (CDCl ): δ = 2.60–2.54 (m, 2
˜
3
H, CH2-N) 2.27 (s, 6 H, N-CH3), 2.26–2.19 (m, 2 H, CH2), 2.15 (s,
6 H, ring CH3), 2.16 (s, 6 H, ring CH3) ppm. The 1H NMR spectro-
side-arm. By employing in situ IR measurements and using scopic data matched the literature values.[21,22]
Preparation of [CpBzIr(CO)2]: A degassed suspension of CpBzTl
(55 mg, 0.154 mmol) and [Ir(CO)3Cl] (21.2 mg, 0.068 mmol) in hex-
iterative band-targeted entropy minimization to analyze the
spectra, however, we have been able to detect the presence
of many hitherto undetected products and intermediates in ane (20 mL) was heated in a Carius tube at 80 °C for 2 d. The
resultant mixture was cooled and filtered through celite to remove
unreacted starting materials. The yellow filtrate was vacuum-dried
the C–H activation reaction of such iridium species, includ-
ing a carboxaldehyde. The formation of the carboxaldehyde
represents the first evidence that the cyclopentadienylirid-
ium system can be photochemically activated to promote
the carbonylation of alkanes.
to give 1e as a yellow oil. IR (cyclohexane): ν = 1969, 2036 cm–1.
˜
1H NMR (CDCl3): δ = 7.23 (m, 5 H, C6H5), 5.45 (m, 2 H, C5H4),
5.37 (m, 2 H, C5H4), 3.77 (s, 2 H, CH2) ppm. The IR data matched
the literature values.[20]
∧
Complex 1c was prepared from Cp Tl (432 mg, 1.265 mmol) and
[Ir(CO)3Cl] (217 mg, 0.696 mmol) in an analogous manner. IR (cy-
Experimental Section
1
clohexane): ν = 1967, 2034 cm–1. H NMR (CDCl ): δ = 5.52 (m,
˜
3
General: All operations were carried out using standard Schlenk
techniques under an inert argon atmosphere unless otherwise
stated. Cyclopentane (99+%), cyclohexane (99.9%), and hexane
were distilled under nitrogen from sodium/benzophenone ketyl.
Chlorotricarbonyliridium (Strem) was purchased and used without
further purification.
2 H, C5H4), 5.37 (m, 2 H, C5H4), 2.58–2.36 (m, 4 H, CH2CH2),
2.26 (s, 6 H, N-CH3) ppm. The 1H NMR spectroscopic data
matched the literature values.[20]
Preparative Photolysis: A solution of the metal complex in the hy-
drocarbon solvent (1 mgmL–1) was placed in a closed quartz tube,
degassed by three cycles of freeze-pump-thaw, and then irradiated
with a 450-W water-cooled, medium-pressure mercury lamp placed
approximately 15 cm away whilst being continuously stirred. For
reaction under CO atmosphere, the quartz tube was refilled with
1 atm of CO from the Schlenk line after degassing. For the laser
irradiation, dichroic mirrors were used to direct the pulses; cooling
by water circulation was not necessary.
NMR spectra were recorded with either a Bruker ACF 300 MHz
or Bruker DPX 300 MHz spectrometer. Chemical shifts are
reported with respect to residual solvent peaks. UV absorption
spectra of starting complexes were recorded in cyclohexane with a
Shimadzu 160 spectrometer. Gas chromatography (GC) analyses
were performed with a HP 6890 gas chromatograph equipped
Eur. J. Inorg. Chem. 2006, 1568–1572
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
1571

P. K. Chan, W. K. Leong, K. I. Krummel, M. V. Garland
FULL PAPER
[4] C. Muller, D. Vos, P. Jutzi, J. Organomet. Chem. 2000, 600,
127–143.
[5] P. Jutzi, T. Redeker, Eur. J. Inorg. Chem. 1998, 663–674.
[6] S. A. Gardner, P. S. Andrews, M. D. Rausch, Inorg. Chem.
1973, 12, 2396–2402.
[7] W. A. G. Graham, J. Organomet. Chem. 1986, 300, 81–91.
[8] D. M. Heinekey, D. A. Fine, T. G. P. Harper, S. T. Michel, Can.
J. Chem. 1995, 73, 1116–1125.
[9] F. P. Pruchnik, K. Waijda-Hermanowicz, M. Koralewicz, J. Or-
ganomet. Chem. 1990, 384, 381–383.
In situ IR Measurements: For measurements at ambient pressure, a
well-stirred solution of 1b or 1d in cyclohexane was circulated be-
tween the quartz reactor and a ZnSe cell through Vitron (internal
diameter 2.06 mm) and Tygon tubing using a peristaltic pump
(MASTERflex C/L 5910 pump systems model 77120–62). UV irra-
diation was carried out through the quartz reactor. The reactor was
fan-cooled and the temperature was not controlled. The atmo-
sphere was either argon or carbon monoxide.
For the high-pressure measurements, a solution of 1b in cyclohex-
ane was circulated between a 100-mL stainless-steel autoclave type
reactor, a high-pressure cell of original design (AMTIR windows),
and an industrial sapphire tube (Almaz Optics, Inc., Marlton, New
Jersey) of 5 mm internal diameter (I.D.) via stainless steel tubing
of 1/16 inch i.d. under various CO pressures. UV irradiation was
carried out through the sapphire tube. The temperature was kept
at 20 °C by cryostatic control.
[10] R. G. Bergman, Science 1984, 223, 902–908.
[11] a) W. Chew, E. Widjaja, M. Garland, Organometallics 2002,
21, 1982–1990; b) E. Widjaja, C. Li, M. Garland, Organometal-
lics 2002, 21, 1991–1997.
[12] We observed no significant difference in product distribution
for the experiments carried out at ambient pressure using Vi-
tron tubing and those carried out under CO pressure using
stainless-steel tubing.
[13] a) T. Sakakura, T. Sodeyama, K. Sasaki, K. Wada, M. Tanaka,
J. Am. Chem. Soc. 1990, 112, 7221–7229; b) G. P. Rosini, K. M.
Zhu, A. S. Goldman, J. Organomet. Chem. 1995, 504, 115–121;
c) G. P. Rosini, W. T. Boese, A. S. Goldman, J. Am. Chem. Soc.
1994, 116, 9498–9505.
Acknowledgments
[14] M. Tanaka, T. Sakukara, in Homogeneous Transition Metal
Catalyzed Reactions (Eds.: W. R. Moser, D. W. Slocum); Ad-
vances in Chemistry Series 230; American Chemical Society:
Washington, DC, 1992, chapter 12, pp 181–196.
[15] F. Gao, K. P. Ng, C. Li, K. I. Krummel, A. D. Allian, M. A.
Garland, J. Catal. 2006, 237, 49–57.
[16] T. F. Wang, T. Y. Lee, J. Organomet. Chem. 1992, 423, 31–38.
[17] P. Jutzi, J. Dahlhaus, Synthesis 1993, 7, 684–686.
[18] P. Singh, M. D. Rausch, J. Organomet. Chem. 1988, 352, 273–
282.
[19] R. G. Ball, W. A. G. Graham, D. M. Heinekey, J. K. Hoyano,
A. D. McMaster, B. M. Mattson, S. T. Michel, Inorg. Chem.
1990, 29, 2023–2025.
[20] M. S. Blais, J. C. W. Chien, M. D. Raush, Organometallics
1998, 17, 3775–3783.
This work was supported by an A*STAR grant (research grant no.
012 101 0035), and K. I. K. and P. K. C. would like to thank the
University for research scholarships.
[1] a) A. H. Janowicz, R. G. Bergman, J. Am. Chem. Soc. 1982,
104, 352–354; b) J. K. Hoyano, W. A. G. Graham, J. Am.
Chem. Soc. 1982, 104, 3723–3725; c) A. H. Janowicz, R. A.
Periana, J. M. Buchanan, C. A. Kovac, J. M. Stryker, M. J.
Wax, R. G. Bergman, Pure Appl. Chem. 1984, 56, 13–23; d)
S. T. Belt, F. W. Grevels, W. E. Klotzbucher, A. McCamley,
R. N. Perutz, J. Am. Chem. Soc. 1989, 111, 8373–8382; e) A. A.
Bengali, R. H. Schultz, C. B. Moore, R. G. Bergman, J. Am.
Chem. Soc. 1994, 116, 9585–9589.
[2] a) J. M. Buchanan, J. M. Stryker, R. G. Bergman, J. Am. Chem.
Soc. 1986, 108, 1537–1550; b) S. E. Bromberg, T. Lian, R. G.
Bergman, C. B. Harris, J. Am. Chem. Soc. 1996, 118, 2069–
2072; c) N. Dunwoody, S. Sun, A. J. Lees, Inorg. Chem. 2000,
39, 4442–4451.
[21] M. S. Blais, M. D. Rausch, J. Organomet. Chem. 1995, 502, 1–
8.
[22] P. Jutzi, M. O. Kristen, B. Neumann, H. G. Stammler, Organo-
metallics 1994, 13, 3854–3861.
Received: November 24, 2005
Published Online: February 28, 2006
[3] A. J. Lees, J. Organomet. Chem. 1998, 554, 1–11.
1572
www.eurjic.org
© 2006 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Inorg. Chem. 2006, 1568–1572