The Complete Delineation of the Initiation, Propagation, and Termination Steps of the Carbomethoxy Cycle for the Carboalkoxylation of Ethene by Pd-Diphosphane Catalysts

Article abstract of DOI:10.1002/anie.200352369

An elegant model: The intermediates in both the hydride and carbomethoxy cycles, and in chain transfer for carboalkoxylation of ethene and CO-alkene copolymerization have been characterized for a highly active cationic palladiumdiphosphane catalyst (see scheme). In contrast to previous reports on model systems, alkene insertion into the Pd-C(O)OMe and Pd-C(O) Me bonds is found to occur at comparable rates.

Full text of DOI:10.1002/anie.200352369

Communications
Reaction-Mechanism Elucidation
The Complete Delineation of the Initiation,
Propagation, and Termination Steps of the
Carbomethoxy Cycle for the Carboalkoxylation of
Ethene by Pd–Diphosphane Catalysts**
Jianke Liu, Brian T. Heaton, Jonathan A. Iggo,* and
Robin Whyman
Drent and Budzelaar have proposed two catalytic cycles for
cationic Pd–diphosphane complex catalyzed CO–alkene
coupling reactions such as CO–ethene copolymerization and
methoxycarbonylation of ethene.[1] The hydride cycle is
initiated by alkene insertion into a Pd–hydride bond whilst
the carbomethoxy cycle is initiated by CO insertion into a
Pd–methoxy bond. The two cycles may operate in parallel, for
example, CO–alkene copolymerization, or one cycle may
dominate, for example, the Lucite process for the methox-
ycarbonylation of ethene to methylpropanoate. Where the
two cycles operate in tandem there exists the possibility of
crossover between the cycles which has been inferred from
chain end-group analysis.[1,2] Product selectivity is determined
in both the initiation and termination step(s).[3–5] The chain-
propagation steps, insertion of CO into a metal–alkyl bond,
and insertion of alkene into a metal–acyl, bond have been
extensively investigated,[6–10] whilst van Leeuwen has recently
discussed termination by alcoholysis of the Pd–alkyl[11] and
Pd–acyl intermediates.[4] We have recently (1) fully charac-
terized the intermediates in and (2) delineated the initiation
process of the hydride cycle.[12–14] The hydride cycle is thus
well established, however, a complete cycle for the Pd–car-
bomethoxy mechanism has not yet been demonstrated
because of the absence of suitable Pd–carbomethoxy pre-
cursors.[15–18] We report herein a simple, general procedure for
the preparation of the key cationic monocarbomethoxy
Pd–diphosphane compounds and present an NMR study
that characterizes, for the first time, all the intermediates in
the carbomethoxy cycle. We also report (1) the delineation of
the propagation steps of the hydride cycle for this catalyst,
and (2) for the first time, characterization of all the complexes
involved in termination of the hydride cycle with concomitant
crossover and initiation of the carbomethoxy cycle. It should
be noted that ligands of the type used in this work give highly
active catalysts in this field of chemistry.[3]
[*] J. Liu, Prof. B. T. Heaton, Dr. J. A. Iggo, Dr. R. Whyman
Department of Chemistry, Donnan and
Robert Robinson Laboratories
University of Liverpool
P. O. Box. 147, Liverpool, (UK) L69 7ZD
Fax: (+44)151-794-3588
E-mail: iggo@liv.ac.uk
[**] The financial support of the EPSRC, grant RG/R37685, is gratefully
acknowledged.
Supporting information for thisarticle isavailable on the WWW
under http://www.angewandte.org or from the author.
90
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/anie.200352369
Angew. Chem. Int. Ed. 2004, 43, 90 –94

Angewandte
Chemie
Table 1: NMR spectroscopic data for the complexes involved in Scheme 1.
The addition of one equivalent
of triethylamine to a CO saturated
solution of [Pd(dibpp)(CH3CN)2]
[CF3SO3]2, (dibpp = 1,3-(iBu2P)2-
C3H6) in a 9:1 mixture of dichloro-
methane and methanol at ꢀ788C,
gives two new doublets at d = 3.7
and, 17.0 ppm, 2J(P,P) = 28 Hz in
Complex
d
J(P,P) J(P,C)
d
31P [ppm] [Hz]
[Hz]
13C [ppm][a]
1 [Pd(dibpp)(OCH3)(CH3CN)][CF3SO3]
3.7 dd 28
17.0
28
3 [Pd(dibpp){C(O)OCH3}(CH3CN)][CF3SO3]
4 [Pd(dibpp){C(O)OCH3}2][b]
8.3 dd 47
ꢀ15.5 dd 47
9
167
191.7 dd
the
31P{1H}
NMR
spectrum
(Table 1). The NMR parameters of
1 are unchanged in the presence of
excess CH3CN, which indicates that
the fourth coordination site in 1 is
occupied by CH3CN rather than
methanol. Complex 1 is therefore
assigned as the Pd–methoxy species
[Pd(dibpp)(OCH3)(CH3CN)]+,
ꢀ8.2
ꢀ8.2
45
45
152
12
203.0
203.0[c]
5 [PdK (dibpp){CH2CH2C(OL )OCH3}][CF3SO3]
6 [Pd(dibpp){CH2CH2C(O)OCH3}(CH3CN)][CF3SO3]
7 [Pd(dibpp){C(O)CH2CH2C(O)OCH3}(CH3CN)][CF3SO3]
8 [PdK (dibpp){C(O)CH2CH2C(OL )OCH3}][CF3SO3]
13.3 d
ꢀ12.0 dd 46
46
193.3 d
10
which is too unstable to be isolated
and reacts readily with 12CO to give
3 (Scheme 1; the putative inter-
mediate [Pd(dibpp)(OCH3)(CO)]+,
2, is not observed, see also
Scheme 2). The 31P{1H} NMR spec-
trum of 3 shows two doublets at d =
8.3 and ꢀ15.5 ppm, 2J(P,P) = 47 Hz.
On using 13CO, these resonances are
further split by 31P–13C coupling,
Table 1, and a corresponding dou-
blet of doublets is observed in the
13C{1H} NMR spectrum at d =
12.7 dd 46
ꢀ12.6 dd 46
2
4
179.0 dd
4.1 dd 68
ꢀ18.6 dd 68
9
116
241.5 dd
173.0 s
17.9 dd 63
ꢀ14.3 dd 63
16
129
228.4 ddd[c]
186.0 d
[a] 13C{1H} NMR spectroscopic data and J(PC) were obtained from experimentsusing 13CO. [b] AA’XX’,
the chemical shifts and coupling constants were obtained from simulation of the experimental spectrum
by using gNMR 4.1. [c] J(CC)=6 Hz.
191.7 ppm,
2J(Ptrans,C) = 167 Hz,
2J(Pcis,C) = 9 Hz, which is character-
istic of the carbonyl of an ester group. Complex 3 can thus be
formulated as [Pd(dibpp){C(O)OCH3}(CH3CN)][CF3SO3].
The addition of a second equivalent of triethylamine leads
to the formation of the Pd bis-carbomethoxy complex
[Pd(dibpp){C(O)OCH3}2], 4. The 31P{1H} and 13C{1H} NMR
spectra of 4, labeled with 13CO, show second order AA’XX’
patterns, which have been simulated. The calculated chemical
shifts and coupling constants for 4 are given in Table 1 and are
similar to those found for 3, thus substantiating the assign-
ment of 3 and 4 as the mono and biscarbomethoxy complexes.
In contrast to the ready reversibility of insertion of CO into
Pd methyl bonds, 3 is stable towards decarbonylation;
solutions of 3 can be taken to dryness in vacuo and extracted
into dichloromethane with no change in the 31P{1H} NMR,
that is, the insertion of CO into the Pd–methoxy bond is
irreversible even though the fourth coordination site is
occupied by CH3CN.[19,20] The dppp (dppp = 1,3-bis(diphenyl-
phosphinyl)propane) analogue can be prepared in a similar
fashion, thus demonstrating the generality of the synthetic
method (see Supporting Information).
The stability of 3 makes it a promising starting point for a
mechanistic NMR study of the carbomethoxy cycle
(Scheme 1). On bubbling ethene through a dichloromethane
solution of 3 at ꢀ788C, followed by warming to ꢀ308C
for 1.5 hr, 31P{1H} and 13C{1H} NMR spectroscopy re-
veals complete conversion of 3 to a new complex 5,
[PdK (dibpp){CH2CH2C(OL )OCH3}]+ (see Table 1). The obser-
vation of a doublet for the ester carbon at d = 193.3 ppm in
the 13C{1H} NMR spectrum, J(P,C) = 10 Hz, indicates that a
five-membered chelating structure is formed in which the
oxygen atom of the ester carbonyl coordinates to the metal
Scheme 1. The “carbomethoxy cycle” for Pd catalyzed CO–ethene cou-
pling.
Angew. Chem. Int. Ed. 2004, 43, 90 –94
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
91

Communications
center. This coupling is lost on addition of acetonitrile to the
solution, which indicates cleavage of the chelate ring and
coordination of CH3CN to the Pd center, 6. There are several
reports of chelating structures, in which the oxygen of an acyl
group coordinates to the Pd centre;[21–24] however, examples
of similar chelating structures involving coordination of the
oxygen atom of the ester group are rare,[11,24,25] and we believe
this is the first example of such a chelating structure for a
cationic diphosphane palladium system formed directly by the
carbomethoxy cycle. A brief report exists of an analogous
dppp compound synthesized by an alternative route.[11]
The reaction of 5 with 13CO at ꢀ308C gives two new Pd
acyl species 7 and 8 that are in equilibrium, Scheme 1. The
13C{1H} NMR spectrum of 7, Table 1, comprises a doublet of
2
doublets at d = 241.5 ppm, J(Ptrans,C) = 116, 2J(Pcis,C) = 9 Hz,
characteristic of an acyl carbonyl group coordinated
to a square planar {Pd(dibpp)} fragment and a singlet
at
d
173 ppm, thus indicating the carbonyl group
of the ester is not coordinated to the Pd center. Com-
plex
complex
Complex
7
can thus be formulated as the “open” acyl
[Pd(dibpp){C(O)CH2CH2C(O)OCH3}(Solv)]+.
Scheme 2. The “hydride” cycle (initiated from
[Pd(dibpp)(CH3)(CH3CN)][CF3SO3] 1’) for Pd-catalyzed CO–ethene cou-
pling. (Primesindicate analogouscomplexesto those in Scheme 1.)
8
is formulated as the “chelating” acyl
[PdK (dibpp){C(O)CH2CH2C(OL )OCH3}]+, on the basis of the
downfield shift of the carbonyl of the ester group, d =
186 ppm, compare with d = 173 ppm in 7. Addition of
methanol to a dichloromethane solution of 5 at room
temperature releases one equivalent of methyl propanoate
as the only organic product detected by 13C{1H} NMR. In the
presence of CO, 3 is regenerated smoothly in situ and can be
used to initiate a new catalytic cycle, Scheme 1.
Base was required to generate the Pd-methoxy species
used in the work above, however, cationic Pd–diphosphane
complex catalyzed CO–alkene coupling reactions such as
CO–ethene copolymerization and methoxycarbonylation of
ethene are normally performed in the presence of an excess of
a strong acid such as CH3SO3H. The same chemistry as that
outlined above can be accessed in the absence of base by
chain crossover from the “hydride” cycle, Scheme 2 and 3.
(Primes are used throughout this work to indicate analogous
intermediates in the cycles shown in Schemes 1 and 2).
Although attempts to prepare the Pd(dibpp)–hydride cation
Scheme 3. Termination of the hydride cycle and crossover to the carbo-
methoxy cycle.
species proceeds via an enolate intermediate but were unable
to characterize the organometallic products of solvolysis of
the enolate.[11] Isomerization of the b-chelate complexes 5’
and 9’ to the enolates 10’a and 10’b, respectively, occurs in the
presence of excess methanol (Scheme 3); in the absence of
methanol decomposition of the b-chelate was observed. The
methoxy species 1 was not observed presumably because of its
instability and readiness to react with CO (see above)[32] to
give 3, from which we have been able to restart the
carbomethoxy cycle described above.
All of the steps in both the hydride and carbomethoxy
catalytic cycles for Pd-catalyzed CO–ethene coupling for a
single, highly active catalyst have thus been fully character-
ized by NMR for the first time. The characterization of the
active catalytic species resulting from chain termination
through protonolysis of the Pd–alkyl intermediate in the
hydride cycle has allowed the pathway from the hydride to the
carbomethoxy cycle through a methanolysis chain-transfer
step to be demonstrated for the first time. Termination of the
carbomethoxy cycle by methanolysis of the alkyl intermediate
also gives 3. It has previously been noted that alcoholysis of
the acyl intermediate gives ester product and a Pd–hydride
were
unsuccessful,
the
methyl
complex
[Pd(dibpp)(CH3)(CH3CN)][CF3SO3] 1’, provides a clean
entry point into the chemistry of the hydride cycle[4–7,11]
(with the exception of the initial insertion of C2H4 into the
ꢀ
Pd H bond). 1’ was prepared in a manner similar to that
described by Brookhart and co-workers.[26] Sequential inser-
tions of CO and ethene were then monitored at low temper-
ature (ꢀ788C to ꢀ308C) by 31P{1H} and 13C{1H} NMR
spectroscopy, and gave the reaction sequence 1’!2’!3’!
5’!7’!9’ (Scheme 2, Table 2) in agreement with reports on
other diphosphane ligands.[6,7,12,14,15,26–31]
Importantly however, we have been able to characterize,
for the first time, all the organometallic species involved in
the methanolysis of the chelate complexes 5’ and 9’, that is,
5’!10a’![1]!3, and 9’!10b’![1]!3 with the exception of
1 (the involvement of which can be inferred from the work
above; see Scheme 3). Van Leeuwen and co-workers have
previously shown that methanolysis of the Pd-chelated alkyl
92
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 90 –94

Angewandte
Chemie
Table 2: NMR spectroscopic data for the complexes involved in Scheme 2 and 3.
Complex
d
31P [ppm]
J(P,P) [Hz]
J(P,C) [Hz]
d
13C [ppm][a]
1’ [Pd(dibpp)(CH3)(CH3CN)][CF3SO3]
11.1 d
ꢀ15.6 d
42
42
2’ [Pd(dibpp)(CO)(CH3)][CF3SO3]
ꢀ0.5 dd
47
47
114
16
181.6 dd
242.6 dd
236.0 d
ꢀ12.3 dd
3’ [Pd(dippp){C(O)CH3}(CH3CN)][CF3SO3]
5.4 dd
ꢀ19.6 dd
70
70
10
112
5’ [PdK (dibpp){CH2CH2C(OL )CH3}][CF3SO3]
12.7 d
ꢀ13.5 dd
46
46
10
7’ [Pd(dippp){(C(O)CH2CH2C(O)CH3)}(CH3CN)][CF3SO3]
9’ [PdK (dibpp){CH2CH2C(OL )CH2CH2C(O)CH3}(CH3CN)][CF3SO3]
10a’ [Pd(dibpp){CH2CH2C(O) CH3}][CF3SO3]
10b’ [Pd(dibpp){CH2CH2C(O)CH2CH2C(O)CH3}][CF3SO3]
3 [Pd(dibpp){C(O)OCH3}(CH3CN)][CF3SO3]
3.9 dd
ꢀ19.1 dd
69
69
10
114
240.3 dd
237.4 dd
208.4 s
14.7 d
ꢀ13.6 dd
47
47
10
12.8 d
ꢀ15.2 d
44
44
12.5 d
ꢀ16.5 d
52
52
206.3 s
7.8 dd
ꢀ15.7 dd
48
48
9
167
190.8 dd
[a] 13C{1H} NMR spectroscopic data and J(PC) were obtained from experimentsby using 13CO.
species “locking” the catalysis into the hydride cycle,[4,33] this
work shows that methanolysis of the alkyl intermediate in
both cycles gives a Pd–methoxy species, thus locking the
catalysis into the carbomethoxy cycle.
[PdK (dibpp){CH2CH2C(OL )CH3}]+ 5’. The chelate ring in 5
can be easily opened by the addition of acetonitrile to a
dichloromethane solution of 5 to give 6 quantitatively.
Complex
6 is formulated as the ring opened cation
Qualitatively, the rate of ethene insertion into the Pd–
carbomethoxy bond in 3 is found to be similar to that of
insertion into the Pd–acyl bond in 3’, that is, insertion occurs
at low temperature on the timescale of tens of minutes. This
contrasts with literature precedent, in which insertion of
alkene into the Pd–carbomethoxy bond is reported not to
occur at room temperature on a timescale of days, if at all.[15,18]
Although we cannot rule out that the high reactivity we
observe is due to the different ligand used—dialkyl versus
diaryl diphosphane—it seems likely that the low reactivity
toward alkene insertion of Pd carbomethoxy complexes
reported in previous work may reflect the presence of a
strongly coordinating anion or ligand in the fourth coordina-
tion site rather than the intrinsic reactivity of the Pd–carbo-
methoxy bond.
[Pd(dibpp){CH2CH2C(O)OCH3}(CH3CN)]+ based on the
upfield shift of the carbonyl carbon (d = 179 ppm) in 6
relative to that in 5, and loss of the 10 Hz P,C coupling
characteristic of a dative bond from the oxygen of the ester
group to the Pd centre. This contrasts with the stability of 5’, in
which the chelating structure is preserved in the presence of
strongly coordinating solvents (e.g. acetonitrile) or anions
(e.g. trifluoroacetate). These results indicate that the ketonic
group is a stronger Lewis base toward the metal centre than
the ester carbonyl group in this system, compare Scheme 1
and 2.
Several six-membered chelates formed by coordination of
the oxygen atom of a ketone to the metal centre have been
reported during the stepwise insertion of alkenes and CO into
metal–acyl compounds[34,35] and we report herein an analo-
gous structure 8 involving coordination of an ester carbonyl
group formed directly from successive insertions of ethene
and CO into a Pd carbomethoxy species. The six-membered
Pd–acyl chelates derived from the hydride cycle are reported
The b-chelate has been proposed to be a resting state in
catalytic cycle,[21] it is therefore interesting to compare the
reactivity of [PdK (dibpp){CH2CH2C(OL )OCH3}]+ 5 with that
of the analogous b-chelate in the hydride cycle
Angew. Chem. Int. Ed. 2004, 43, 90 –94
www.angewandte.org
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
93

Communications
[15] G. Dekker, C. J. Elsevier, K. Vrieze, P. W. N. M. van Leeuwen,
C. F. Roobeek, J. Organomet. Chem. 1992, 430, 357.
[16] A. Dervisi, P. G. Edwards, P. D. Newman, R. P. Tooze, S. J. Coles,
M. B. Hursthouse, J. Chem. Soc. Dalton Trans. 1999, 1113.
[17] M. Sperrle, G. Consiglio, J. Mol. Catal. A 1999, 143, 263.
[18] G. Cavinato, L. Toniolo, J. Organomet. Chem. 1990, 398, 187.
[19] J. E. Byrd, J. Halpern, J. Am. Chem. Soc. 1971, 93, 1634.
[20] H. C. Clark, K. R. Dixon, W. J. Jacobs, J. Am. Chem. Soc. 1969,
91, 1346.
[21] W. P. Mul, H. Oosterbeek, G. A. Beitel, G. J. Kramer, E. Drent,
Angew. Chem. 2000, 112, 1919; Angew. Chem. Int. Ed. 2000, 39,
1848.
[22] M. J. Green, G. J. P. Britovsek, K. J. Cavell, F. Gerhards, B. F.
Yates, K. Frankcombe, B. W. Skelton, A. H. White, J. Chem. Soc.
Dalton Trans. 1998, 1137.
[23] P. Braunstein, C. Frison, X. Morise, Angew. Chem. 2000, 112,
2989; Angew. Chem. Int. Ed. 2000, 39, 2867.
[24] S. Mecking, L. K. Johnson, L. Wang, M. Brookhart, J. Am.
Chem. Soc. 1998, 120, 888.
[25] E. G. Lundquist, K. Folting, W. E. Streib, J. C. Huffman, O.
Eisenstein, K. G. Caulton, J. Am. Chem. Soc. 1990, 112, 855.
[26] C. S. Shultz, J. Ledford, J. M. DeSimone, M. Brookhart, J. Am.
Chem. Soc. 2000, 122, 6351.
[27] I. del Rio, C. Claver, P. W. M. N. van Leeuwen, Eur. J. Inorg.
Chem. 2001, 2719.
to promote methanolysis relative to further ethene inser-
tion,[3] however, our preliminary results show that further
ethene insertion into 7 is preferred over methanolysis. These
findings indicate that the oxochelates (5 and 8) formed in the
carbomethoxy cycle are more susceptible to nucleophilic
attack by substrates such as ethene, strong coordinating
solvents, and anions, than their counterparts in the hydride
cycle.
A Pd–carbomethoxy compound 3 has been synthesized by
a simple methodology; this has allowed the Pd carbomethoxy
cycle in CO–ethene coupling to be demonstrated for the first
time and all important intermediates in both the carbome-
thoxy and hydride cycles to be characterized by 31P and
13C NMR spectroscopy. In contrast to previous reports,
comparable reactivity towards ethene insertion was observed
for the Pd carbomethoxy complex 3 and the Pd acetyl
complex 3’ and is attributed to the absence of a strongly
coordinating ligand in the fourth coordination site of 3. In
copolymerization and methoxycarbonylation reactions, the
chain termination steps are as important as initiation and
propagation since they can govern the product selectivity and
regeneration of the catalytically active species. Methanolysis
of the Pd b-chelate complexes occurs in both cycles regen-
[28] K. Vrieze, J. H. Groen, J. G. P. Delis, C. J. Elsevier, P. W. N. M.
van Leeuwen, New J. Chem. 1997, 21, 807.
[29] C. Bianchini, A. Meli, G. Muller, W. Oberhauser, E. Passaglia,
Organometallics 2002, 21, 4965.
[30] W. Clegg, G. R. Eastham, M. R. J. Elsegood, B. T. Heaton, J. A.
Iggo, R. P. Tooze, R. Whyman, S. Zacchini, Organometallics
2002, 21, 1832.
ꢀ
erating the Pd OCH3 species 1 which, in the presence of CO,
ꢀ
forms the Pd C(O)OCH3 species initiating a new catalytic
cycle.
Received: July 14, 2003
Revised: October 6, 2003 [Z52369]
[31] J. Wolowska, G. R. Eastham, B. T. Heaton, J. A. Iggo, C. Jacob,
R. Whyman, Chem. Commun. 2002, 2784.
[32] D. Milstein, Acc. Chem. Res. 1988, 21, 428.
[33] R. I. Pugh, E. Drent, Catalytic Synthesis of Alkene-Carbon
Monoxide Copolymers and Cooligomers, Vol. 27 (Ed.: A. Sen),
Kluwer, Dordrecht, 2003, chap. 2.
[34] R. A. M. Robertson, A. D. Poole, M. J. Payne, D. J. Cole-
Hamilton, Chem. Commun. 2001, 47.
Keywords: carbonylation · copolymerization · homogeneous
catalysis · palladium · polyketones · reaction mechanisms
.
[1] E. Drent, P. H. M. Budzelaar, Chem. Rev. 1996, 96, 663.
[2] Catalytic Synthesis of Alkene-Carbon Monoxide Copolymers and
Cooligomers, Vol. 27 (Ed.: A. Sen), Kluwer, Dordrecht, 2003.
[3] R. I. Pugh, E. Drent, Adv. Synth. Catal. 2002, 344, 837.
[4] P. W. N. M. van Leeuwen, M. A. Zuideveld, B. H. G. Swennen-
huis, Z. Freixa, P. C. J. Kamer, K. Goubitz, J. Fraanje, M. Lutz,
A. L. Spek, J. Am. Chem. Soc. 2003, 125, 5523.
[35] F. C. Rix, M. Brookhart, P. S. White, J. Am. Chem. Soc. 1996, 118,
4746.
[5] Z. Freixa, P. W. N. M. van Leeuwen, J. Chem. Soc. Dalton Trans.
2003, 1890.
[6] J. Ledford, C. S. Shultz, D. P. Gates, P. S. White, J. M. DeSimone,
M. Brookhart, Organometallics 2001, 20, 5266.
[7] K. Nozaki, N. Sato, Y. Tonomura, M. Yasutomi, H. Takaya, T.
Hiyama, T. Matsubara, N. Koga, J. Am. Chem. Soc. 1997, 119,
12779.
[8] C. Bianchini, A. Meli, Coord. Chem. Rev. 2002, 225, 35.
[9] G. Consiglio, Chimia 2001, 55, 809.
[10] C. Bianchini, A. Meli, W. Oberhauser, J. Chem. Soc. Dalton
Trans. 2003, 2627.
[11] M. A. Zuideveld, P. C. J. Kamer, P. W. M. N. van Leeuwen,
P. A. A. Klusener, H. A. Stil, C. F. Roobeek, J. Am. Chem. Soc.
1998, 120, 7977.
[12] G. R. Eastham, B. T. Heaton, J. A. Iggo, R. P. Tooze, R.
Whyman, S. Zacchini, Chem. Commun. 2000, 609.
[13] G. R. Eastham, R. P. Tooze, M. Kilner, D. F. Foster, D. J. Cole-
Hamilton, J. Chem. Soc. Dalton Trans. 2002, 1613.
[14] W. Clegg, G. R. Eastham, M. R. J. Elsegood, B. T. Heaton, J. A.
Iggo, R. P. Tooze, R. Whyman, S. Zacchini, J. Chem. Soc. Dalton
Trans. 2002, 3300.
94
ꢀ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
Angew. Chem. Int. Ed. 2004, 43, 90 –94