A unique Pd4 platform with CH3 and μ-CH 2 groups and its C-C coupling reaction with simple olefins

Article abstract of DOI:10.1002/anie.200904745

(Chemical Equation Presented) Menage a quatre: The Pd ₄ complex 1 features both terminal CH₃ and bridging CH₂ groups, and it reacts with ethylene at room temperature to give mainly propene. NMR spectroscopic studies reveal several intermediates in the formation of 1 from Pd₂ building blocks.

Full text of DOI:10.1002/anie.200904745

Communications
DOI: 10.1002/anie.200904745
Multinuclear Complexes
À
A Unique Pd4 Platform with CH3 and m-CH2 Groups and Its C C
Coupling Reaction with Simple Olefins**
Anna Sachse, Michael John, and Franc Meyer*
Bi- and oligonuclear complexes currently attract significant
attention in the chemical community, mainly because coop-
erative effects of adjacent metal ions are expected to lead to
unique substrate activation modes and to novel reactivity
patterns.[1] While much of the work in this field is inspired by
natural enzymes that contain multinuclear active sites,[2]
Scheme 1. Pyrazolate-based dipalladium(II) complex A.
recent years have also seen growing interest and impressive
progress in the development of non-biomimetic binuclear
complexes for two-center organometallic catalysis,[3,4] includ-
ing the homogeneous polymerization of olefins.[5,6] An
interesting situation arises for multinuclear complexes that
contain more than two metal ions, as little is known about the
synergetic effects in such systems.
Compartmental ligand systems are of pivotal importance
for achieving and controlling metal-ion cooperativity.[7] They
allow defined positioning of two or more metal ions and may
support controlled aggregation to yield high-nuclearity com-
plexes. In this regard pyrazolate-derived bridging ligands are
very useful scaffolds, as their decoration with chelating side
arms in the 3- and 5-positions of the central heterocycle
provides a means of deliberately tuning various character-
istics of the binuclear array.[8] We have recently introduced a
series of pyrazole-based ligands that possess appended imine
donor functions with bulky aryl substituents.[6,9] These systems
can be described as having two adjacent binding compart-
ments akin to the a-diimine type, and their bi- or oligonuclear
palladium(II) and nickel(II) complexes, after activation with
methylalumoxane (MAO), indeed serve as highly active
catalysts for olefin polymerization.[6]
We were interested in elucidating details of the activation
process as well as possible internuclear effects in the activated
species. Therefore we investigated the methylation of a
representative pyrazolate-based dipalladium(II) complex A
(Scheme 1)[6a] using relatively mild agents such as SnMe4.
Unexpectedly, we observed the formation of an unprece-
dented tetrapalladium complex (1, Scheme 2) that features
both terminal CH3 and m-CH2 groups, and herein we report
Scheme 2. Proposed reaction sequence leading to 1.
[*] Dr. A. Sachse, Dr. M. John, Prof. Dr. F. Meyer
Institut fꢀr Anorganische Chemie
Georg-August-Universitꢁt Gꢂttingen
Tammannstrasse 4, 37077 Gꢂttingen (Germany)
The reaction of A in dichloromethane with an excess of
Fax: (+49)551-39-3063
E-mail: franc.meyer@chemie.uni-goettingen.de
insight into its formation, its molecular structure, and its
intriguing reactivity with simple olefins such as ethylene.
SnMe4 results in the formation of a yellow solution, from
which yellow crystals of complex 1 were obtained in good
yields at À208C.[10] The molecular structure of 1 was
determined by single-crystal X-ray diffraction and is shown
in Figure 1. Selected bond lengths and angles are listed in
Table S1; crystallographic data are summarized in Table S2;
Homepage: http://www.meyer.chemie.uni-goettingen.de
[**] Financial support from the Fonds der Chemischen Industrie is
gratefully acknowledged.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.200904745.
1986
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 1986 –1989

Angewandte
Chemie
In a reaction of A with approximately 30 equiv SnMe4,
performed at NMR scale, it was possible to detect two
sequential intermediates en route to 1. The first (intermediate
I1) has an asymmetric structure (Cs), while the second
(intermediate I2) is symmetric (C2v). NOESY correlations of
the isopropyl methyl groups show the presence of palladium-
bound methyl groups in both intermediates (part of the
NOESY spectrum is depicted in the Supporting Information,
Figure S3). To distinguish the intermediates by their molec-
ular size, diffusion coefficients were measured using DOSY
NMR (Figure S4 in the Supporting Information). Diffusion
coefficients D of intermediates I1 and I2 ((1.14 Æ 0.03) ꢁ
10À9 m2 sÀ1) are substantially higher than that of 1 ((0.85 Æ
0.03) ꢁ 10À9 m2 sÀ1) but in the same range as that of the starting
material A ((1.10 Æ 0.03) ꢁ 10À9 m2 sÀ1). We thus assume a
binuclear {LPd2} motif for both intermediates.
Figure 1. Molecular structure of 1. For clarity, all hydrogen atoms are
omitted.
The buildup and decay of the different intermediates and
products was monitored over time by NMR spectroscopy
(Figure 2). The concentrations of intermediates I1 and I2 reach
their maxima after 6 and 10 h, respectively, while 1 becomes
detectable after approximately 5 h and predominates after
an ORTEP plot is provided in Figure S1 in the Supporting
Information.[11]
Complex 1 contains two {LPd2} subunits (L = pyrazolate
ligand) that have assembled to give a Pd4 platform with two
terminal CH3 groups (C33/C33’ at the outer Pd1/Pd1’
centers), a bridging CH2 group (C34 spanning the central
metal atoms Pd2···Pd2’), and two bridging chlorides (Cl1/Cl1’
linking the two subunits along the edges Pd1···Pd2’ and
Pd1’···Pd2). Metal–metal separations in the pyrazolate-
bridged binuclear constituents {LPd2} are much longer
(4.4 ꢀ) than in the starting complex A (3.8 ꢀ), which goes
along with a severe twisting of the two Pd coordination planes
in each {LPd2} unit against each other by 528 (N1/N3/C33/Cl1/
Pd1 and N2/N4/Cl1’/C34/Pd2). The two {LPd2} subunits are
tilted by 698 with respect to each other, and the m-CH2 ligand
is associated with a Pd2···Pd2’ separation of 3.3 ꢀ and a Pd2-
C34-Pd2’ angle of 1098. All palladium atoms have four-
coordinate, distorted square-planar coordination spheres.
À
While mononuclear palladium complexes with Pd CH3
bonds are quite abundant and well-studied,[12] there are
Figure 2. Buildup and decay of selected intermediates and products in
the 1H NMR spectrum of the reaction of 1 with approximately 30 equiv
relatively few examples of binuclear palladium complexes
with several Pd CH3 bonds. The Pd-(m-CH2)-Pd motive is
even less common; only two examples are known, namely
1
2
&
[13]
of SnMe4.
complex A, 3 intermediate I , ? intermediate I ,
H
^
À
!
~
complex 1, product 2, methane, ethane.
[Pd2Cl2(m-CH2)(m-dppm)2] (dppm = bis(diphenylphosphino)-
methane)[14]
and
[(N^N)Pd(m-CH2)(m-Me)Pd(N^N)]+,
[15]
=
=
(N^N = (C6H3-iPr2-2,6)-N CMeCMe N-(C6H3-iPr2-2,6)).
13 h, when less than 10% of the starting material remains.
Formation of SnMe3Cl is obvious both from 119Sn (d =
168 ppm) and 1H NMR spectroscopy (d = 0.66 ppm). Con-
comitant with the buildup of 1, the formation of methane
À
The latter contains a Pd Pd bond that results in a much
shorter metal–metal separation (2.7 ꢀ) and a more acute Pd-
(m-CH2)-Pd angle (858), while the former exhibits parameters
similar to 1 (d(Pd···Pd) = 3.2 ꢀ, ](Pd-(m-CH2)-Pd) = 1038).
However, 1 is the first Pd complex that contains both terminal
CH3 and m-CH2 groups.
1
could be detected by H NMR spectroscopy (d = 0.21 ppm)
and was confirmed by DOSYexperiments. At longer reaction
times a second (as yet unidentified) product slowly appears,
which is accompanied by the formation of ethane. This later
product 2 only forms if an excess of SnMe4 is present, and its
formation is highly dependent on the SnMe4 concentration.
The wealth of information from NMR spectroscopy
allows us to sketch a reaction sequence for the formation of
1 from A (Scheme 2). Complex A is sequentially methylated
by SnMe4 to form intermediates I1 and I2, with subsequent
dimerization of two molecules I2 and the elimination of CH4.
Complex 1 was fully characterized by NMR spectroscopy
using 1D and 2D 1H and 13C techniques. The 1H NMR
spectrum of 1 in CD2Cl2 shows the expected splitting of the
isopropyl methyl group (CH3iPr) signals into eight doublets
and a splitting of the isopropyl methine group (CHiPr) signals
into four separate septets (3JHH = 6.8 Hz; 1H NMR spectra of
1 and A are given in the Supporting Information, Figure S2.)
An upfield signal at d = À0.06 ppm can be assigned to the
methyl groups on the palladium atoms, while the m-CH2
resonance is detected at d = 3.8 ppm.
À
Mechanistically, formation of a m-CH2 ligand from two Pd
CH3 fragments has been proposed to proceed by activation of
Angew. Chem. Int. Ed. 2010, 49, 1986 –1989
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1987

Communications
+
À
a methyl C H bond at a cationic [MePdLn] species, with
subsequent hydride transfer to Pd and reductive elimination
of methane.[15,16] In the present case, an electrophilic metal
but are formed independently. When propene was employed
as a substrate, trans-2-butene, cis-2-butene, and 2-methylpro-
pene were detected, although in this case the reaction was at
least a factor of five slower. To determine the origin of the
C atoms in the C3 and C4 products, 13C-labeled ethylene
(13C2H4) was used under the same conditions, and the
isotopomers of the different product olefins were identified
by 13C NMR spectroscopy (Scheme 3, Figure S6 in the
Supporting Information). trans-2-Butene, cis-2-butene, and
1-butene were found to be fully 13C-labeled, which indicates
site may possibly be generated by a transient shift of the m-Cl
2
À
ligand in I to a nonbridging position and intermolecular C H
activation according to the situation in I3 (Scheme 2).
Besides its unique structure, 1 is of interest since terminal
CH3 and bridging CH2 groups are considered as essential
components for various catalytic transformations on metal
clusters and surfaces,[17] and oligonuclear species have been
suggested as (often unobserved) resting states in catalytic
systems.[18] Since 1 contains both CH3 and m-CH2 groups in
close proximity on a tetranuclear platform, we investigated its
reactivity towards simple olefins. When ethylene was bubbled
through a CD2Cl2 solution of 1 at room temperature, propene
was detected as the major product (by monitoring with
1H NMR spectroscopy), together with some trans-2-butene,
cis-2-butene, 1-butene, and ethane as well as traces of
methane (Figure S5 in the Supporting Information); forma-
tion of these compounds was also confirmed by GC-MS.
Gradual deposition of Pd0 and reconstitution of some A
occurred during the reaction, thus indicating gradual decom-
position of the resulting organopalladium complexes. The
time course of product accumulation was monitored by
1H NMR spectroscopy, which clearly revealed the emergence
of the characteristic olefin resonances and concomitant
Scheme 3. Olefins resulting from the reaction of 1 with 13C2H4;
propene isotopomers are the major products.
that all C4 olefins result from the dimerization of two ethylene
molecules. In contrast, the major product propene contained
only two 13C labels and is thus assembled from one molecule
of ethylene and an unlabeled C1 unit from complex 1 (either
CH3 or m-CH2). The minor byproduct ethane was fully 13C-
labeled (suggesting that it results from hydrogenation of the
13C2H4 substrate), while no 13C is found in the trace amounts
of methane.
À
disappearance of the Pd Me and m-CH2 signals (Figure 3).
Given the triple bridging of the two {LPd2} subunits by m-CH2
and two m-Cl ligands, it appears unlikely that 1 initially
13
13
13
À
=
À
=
Propene isotopomers H3
C
CH CH2 and H3C CH
13CH2 are formed in a 1:1 ratio right from the beginning,
while 1-butene is gradually isomerized to the more stable 2-
butenes (Figure 3), thus suggesting that H-scrambling can
À
occur, presumably by Pd-mediated C H activation. This
scrambling is confirmed by experiments using
1 and
[D4]ethylene under the same conditions. GC-MS of the gas
phase above the reaction mixture using an SPME (solid phase
microextraction) probe as well as NMR spectroscopic mon-
itoring of the solution (an example of an 1H{2H} NMR
spectrum is depicted in Figure S7 in the Supporting Informa-
tion) showed that all possible isotopomers of propene and
butenes are present. We were also able to detect labeled
À
methane (CH3D), ethane isotopomers (C2D5H, CD2H
CD2H), and [D2]ethylene. The ratio of the different propene
and butene isotopomers changes over time, gradually accu-
mulating those with higher D content. This accumulation
Figure 3. Decay of 1 and buildup of product olefins in the reaction
~
&
*
with 22 equiv ethylene. complex 1, propene, trans-2-butene,
3 cis-2-butene, ? 1-butene.
apparently results from continuous H/2H exchanges involv-
1
ing the [D4]ethylene substrate.
In conclusion, a novel Pd4 complex has been discovered
that contains both CH3 and m-CH2 groups in close proximity
on the tetranuclear platform, and several intermediates have
been spectroscopically identified during its formation from
binuclear precursors. This unique Pd4 system reacts with
ethylene to give mainly propene (and some butenes), which
has been investigated by isotopic labeling. Besides the
interesting structural motif, the present results corroborate
that oligonuclear species may well be involved, though often
not detected, in reactions with organopalladium catalysts.
Once cooperative reactivity patterns are better understood,
oligonuclear sites will open new pathways for substrate
converts to binuclear fragments upon reaction with ethylene.
Furthermore, two lines of evidence suggest that transforma-
tion of ethylene is mediated by 1 itself, not by any later
decomposition products: 1) propene and butenes are gener-
ated right from the beginning without any initiation phase,
and 2) their formation ceases when 1 has disappeared
(Figure 3). It should also be noted that complex A is
completely unreactive towards ethylene.
The simultaneous rise of propene and butenes suggests
that the latter do not stem from the primary product propene
1988
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 1986 –1989

Angewandte
Chemie
J.-L. Pierre, Eur. J. Inorg. Chem. 2003, 4137 – 4146; d) F. Meyer,
transformations. Current work focuses on elucidating initial
Eur. J. Inorg. Chem. 2006, 3789 – 3800; e) M. Jarenmark, H.
Carlsson, E. Nordlander, C. R. Chim. 2007, 10, 433 – 462.
[3] See, for example: a) C. Dubs, T. Yamamoto, A. Inagaki, M.
Akita, Organometallics 2006, 25, 1359 – 1367; b) B. M. Trost, S.
Hisaindee, Org. Lett. 2006, 8, 6003 – 6005; c) D. J. Anderson, R.
McDonald, M. Cowie, Angew. Chem. 2007, 119, 3815 – 3818;
Angew. Chem. Int. Ed. 2007, 46, 3741 – 3744; d) M. F. Pilz, C.
Limberg, B. B. Lazarov, K. C. Hultzsch, B. Ziemer, Organo-
metallics 2007, 26, 3668 – 3676; e) D. C. Powers, T. Ritter, Nat.
Chem. 2009, 1, 302 – 309.
[4] Multimetallic Catalysts in Organic Synthesis (Eds.: M. Shibasaki,
Y. Yamamoto), Wiley-VCH, Weinheim, 2004.
[5] H. Li, T. J. Marks, Proc. Natl. Acad. Sci. USA 2006, 103, 15295 –
15302.
[6] a) G. Noꢃl, J. C. Rꢄder, S. Dechert, H. Pritzkow, L. Bolk, S.
Mecking, F. Meyer, Adv. Synth. Catal. 2006, 348, 887 – 897; b) A.
Sachse, V. Daebel, A. Lange, S. Demeshko, S. Dechert, F. Meyer,
Dalton Trans. 2010, DOI: 10.1039/b925535d.
[7] L. Gavrilova, B. Bosnich, Chem. Rev. 2004, 104, 349 – 383.
[8] J. Klingele, S. Dechert, F. Meyer, Coord. Chem. Rev. 2009, 253,
2698 – 2741.
[9] a) J. C. Rꢄder, F. Meyer, M. Konrad, S. Sandhꢄfner, E. Kaifer, H.
Pritzkow, Eur. J. Org. Chem. 2001, 4479 – 4487; b) J. C. Rꢄder, F.
Meyer, H. Pritzkow, Chem. Commun. 2001, 2176 – 2177; c) A.
Sachse, G. Noꢃl, S. Dechert, S. Demeshko, A. Honecker, A.
Alfonsov, V. Kataev, F. Meyer, Eur. J. Inorg. Chem. 2008, 5390 –
5396; d) A. Sachse, S. Demeshko, F. Meyer, Dalton Trans. 2009,
7756 – 7764.
[10] When only a stoichiometric amount of SnMe4 is used, the
reaction is very slow and does not give a single product.
[11] CCDC 743783 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from
The Cambridge Crystallographic Data Centre via www.ccdc.
cam.ac.uk/data_request/cif.
steps and further details of the interaction of olefins with 1.
Experimental Section
General considerations: All NMR experiments were performed at
258C on a Bruker Avance 500 MHz spectrometer using standard
parameters. 1H and 13C chemical shifts were calibrated to the internal
solvent signals (d = 5.32 and 53.8 ppm for CD2Cl2) and peaks were
assigned using 2D experiments: COSY, NOESY (500 ms mixing),
CH-COSY/HSQC, and HMBC (optimized for JCH = 7 Hz). DOSY
spectra were recorded with 2 ms bipolar z-gradient pulses ramped
1
linearly from 1 to 50 GcmÀ1 and a diffusion delay of 70 ms. H{2H}
(2.5 kHz decoupling) and 2H NMR experiments were performed with
the lock channel switched off. 119Sn chemical shifts were calibrated
with the SnMe4 signal. IR spectra from KBr pellets were recorded on
a
Digilab Excalibur Series FTS 3000 spectrometer. Elemental
analyses were carried out with a Heraeus CHN-O-RAPID instru-
ment. Complex A was prepared as reported;[6] all other reagents were
purchased from commercial sources and used as supplied.
1: SnMe4 (1.73 mL, 12.4 mol) was added to a solution of A
(250 mg, 0.31 mol) in CH2Cl2 (100 mL). The reaction mixture was
stirred at room temperature for 15 h. After evaporation of the solvent
and excess SnMe4 under reduced pressure, the crude product was
redissolved in CH2Cl2 and filtered over Celite. The pure product was
obtained after evaporation of the solvent and recrystallization from
CH2Cl2 at À208C. Yellow crystals; yield: 200 mg (86%). IR (KBr):
n˜ = 3441 (br), 3063 (w), 2960 (vs), 2925 (s), 2868 (m), 1708 (m), 1626
(m), 1564 (vs), 1459 (m), 1435 (vs), 1383 (m), 1363 (m), 1326 (m), 1260
(m), 1225 (m), 1189 (w), 1158 (w), 1098 (m), 1058 (m), 1023 (w), 967
(w), 936 (w), 868 (w), 843 (w), 800 (m), 771 (m), 580 (w), 530 (w) cmÀ1
.
1H NMR (500 MHz, CD2Cl2): d = À0.06 (s, 6H, MePd), 1.06–1.55 (m,
48H, MeiPr), 2.04 (s, 6H, Meimine), 2.12 (s, 6H, Meimine), 2.49 (s, 6H,
Mepz4), 2.71 (sept, 3JHH = 6.8 Hz, 2H, CHiPr), 2.92 (sept, 3JHH = 6.8 Hz,
3
3
2H, CHiPr), 3.28 (sept, JHH = 6.8 Hz, 2H, CHiPr), 3.52 (sept, JHH
=
À
À
[12] A search of the CSD for Pd CH3 units gave 336 hits with d(Pd
6.8 Hz, 2H, CHiPr), 3.80 (s, 2H, m-CH2), 7.07–7.16 ppm (m, 12H,
CHAr). 13C NMR (125 MHz, CD2Cl2): d = 2.4 (MePd), 11.5 (Mepz4),
20.0, 20.9 (Meimine), 22.9, 23.5, 23.9, 23.9, 24.1, 24.2, 24.2, 25.0 (MeiPr),
28.2, 28.3, 28.5, 28.7 (CHiPr), 50.7 (m-CH2), 120.5 (Cpz4), 123.4, 123.8,
124.0, 125.9, 127.1 (CHAr), 140.0, 140.5, 140.8, 141.0, 141.7, 142.5 (CAr),
150.5, 153.3 (Cpz3,5), 165.3, 170.3 ppm (Cimine). UV/Vis (CH2Cl2): lmax
(e/LmolÀ1 cmÀ1) = 245 (26), 275 (27), 313 nm (26). Elemental analysis
(%) calcd for C67H94N8Cl2Pd4: C 53.36, H 6.28, N 7.43; found:
C 53.74, H 6.24, N 7.39.
C) in the range from 1.946 to 2.309 ꢀ, a mean value of 2.057 ꢀ,
À
and a median value of 2.046 ꢀ; d(Pd1 C33) in 1 is 2.036(8) ꢀ.
[13] See, for example: a) S. J. Young, B. Kellenberger, J. H. Reibens-
pies, S. E. Himmel, M. Manning, O. P. Anderson, J. K. Stille, J.
Am. Chem. Soc. 1988, 110, 5744 – 5753; b) A. Satake, H.
Koshino, T. Nakata, J. Organomet. Chem. 2000, 595, 208 – 214;
c) C. R. Baar, M. C. Jennings, R. J. Puddephatt, Organometallics
2001, 20, 3459 – 3465; d) T. Rꢂther, M. C. Done, K. J. Cavell, E. J.
Peacock, B. W. Skelton, A. H. White, Organometallics 2001, 20,
5522 – 5531; e) A. D. Burrows, M. F. Mahon, S. P. Nolan, M.
Varrone, Inorg. Chem. 2003, 42, 7227 – 7238; f) A. Scarel, J.
Durand, D. Franchi, E. Zangrando, G. Mestroni, C. Carfagna, L.
Mosca, R. Seraglia, G. Consiglio, B. Milani, Chem. Eur. J. 2005,
11, 6014 – 6023; g) N. M. Motsoane, I. A. Guzei, J. Darkwa, Z.
Naturforsch. B 2007, 62, 323 – 330.
Received: August 25, 2009
Revised: October 27, 2009
Published online: February 12, 2010
À
Keywords: alkene coupling · bridging ligands · C C coupling ·
multinuclear complexes · palladium
.
[14] S. R. Klopfenstein, C. Kluwe, K. Kirschbaum, J. A. Davies, Can.
J. Chem. 1996, 74, 2331 – 2339.
[15] J. H. Brownie, M. C. Baird, Organometallics 2003, 22, 33 – 41.
[16] M. Bochmann, T. Cuenca, D. T. Hardy, J. Organomet. Chem.
1994, 484, c10 – c12.
[17] a) J. R. Torkelson, R. McDonald, M. Cowie, J. Am. Chem. Soc.
1998, 120, 4047 – 4048; b) P. M. Maitlis, J. Organomet. Chem.
2004, 689, 4366 – 4374.
[18] See, for example: C. Markert, M. Neuburger, K. Kulicke, M.
Meuwly, A. Pfaltz, Angew. Chem. 2007, 119, 5996 – 5999; Angew.
Chem. Int. Ed. 2007, 46, 5892 – 5895.
[1] a) G. Sꢂss-Fink, Angew. Chem. 1994, 106, 71 – 73; Angew. Chem.
Int. Ed. Engl. 1994, 33, 67 – 69; b) E. K. van den Beuken, B.
Feringa, Tetrahedron 1998, 54, 12985 – 13011; c) B. Bosnich,
Inorg. Chem. 1999, 38, 2554 – 2562; d) S. Matsunaga, M. Shiba-
saki, Bull. Chem. Soc. Jpn. 2008, 81, 60 – 75.
[2] a) H. Steinhagen, G. Helmchen, Angew. Chem. 1996, 108, 2489 –
2492; Angew. Chem. Int. Ed. Engl. 1996, 35, 2339 – 2342; b) D. E.
Fenton, H. Okawa, Chem. Ber. 1997, 130, 433 – 442; c) C. Belle,
Angew. Chem. Int. Ed. 2010, 49, 1986 –1989
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
1989