Nickel-catalyzed regioselective [2+2+2] cycloaddition of carboryne with alkynes

Article abstract of DOI:10.1002/anie.201001249

(Represented Chemical Equation) A View [2+2+2] a Kill: For the first time, nickel-catalyzed [2+2+2] cycloaddition reactions of alkynes or diynes with carboryne using 1-iodo-2-lithiocarborane as a precursor have been achieved. The mechanism is proposed after the structural confirmation of the key intermediate, nickelacyclopentene.

Full text of DOI:10.1002/anie.201001249

Angewandte
Chemie
DOI: 10.1002/anie.201001249
Cycloaddition Reactions
Nickel-Catalyzed Regioselective [2+2+2] Cycloaddition of Carboryne
with Alkynes**
Zaozao Qiu, Sunewang R. Wang, and Zuowei Xie*
Carboryne (1,2-dehydro-ortho-carborane), a three-dimen-
sional relative of benzyne, was first reported in 1990 as a
highly reactive intermediate.[1] Subsequent studies of its
reactivity showed that it can react with alkenes, dienes, and
alkynes in [2+2] and [2+4] cycloaddition, and ene-reaction
patterns,[2] similar to that of benzyne. The carboryne reactions
are usually complicated and do not proceed in a controlled
manner. On the other hand, nickel–carboryne complex [(h2-
C2B10H10)Ni(PPh3)2][3] can undergo regioselective [2+2+2]
cycloaddition reactions with 2 equivalents of alkyne to afford
benzocarboranes,[4] can react with 1 equivalent of alkenes to
generate alkenylcarborane coupling products,[5] and can
undergo a three-component [2+2+2] cyclotrimerization reac-
tion with 1 equivalent of activated alkene and 1 equivalent of
alkyne to give dihydrobenzocarboranes.[6] However, these
reactions require a stoichiometric amount of nickel reagent.
In view of the analogy between metal–benzyne and metal–
carboryne complexes[7,8] and the metal-catalyzed reactions of
benzyne with alkenes and alkynes,[9] we wondered if a
catalytic version of these nickel-mediated carboryne reactions
could be developed.
1-Iodo-2-lithiocarborane was conveniently prepared
in situ from the reaction of dilithiocarborane with 1 equiv-
alent of iodine in toluene at room temperature; importantly,
1-iodo-2-lithiocarborane was much more thermally stable
than 1-bromo-2-lithiocarborane. Heating a solution of 1-iodo-
2-lithiocarborane in benzene overnight afforded the [4+2]
cycloaddition product 1,2-(2,5-cyclohexadiene-1,4-diyl)-
ortho-carborane in 25% yield, much higher than the 8%
yield that is afforded from the 1-bromo-2-lithiocarborane
precursor.[10] This result suggests that 1-iodo-2-lithiocarbor-
ane is a more efficient precursor than the bromo one. We then
examined the catalytic activity of various metal complexes in
the reaction of 1-iodo-2-lithiocarborane with an excess
amount of 3-hexyne in toluene at 1108C for 2 hours and the
results are summarized in Table 1. The Ni0 complexes were all
catalytically active with [Ni(cod)2] (cod = 1,5-cyclooctadiene)
being the most active, giving the desired [2+2+2] cyclo-
addition product 2a in 33–49% yield (Table 1, entries 1–3).
Addition of PPh3 led to a big drop in the yield of 2a from 49%
to 33%, presumably because free PPh3 molecules compete
with the alkyne for the coordination site on the nickel atom.
We learnt from the previous stoichiometric reactions that
high temperatures were necessary for the insertion of alkynes
Table 1: Optimization of reaction conditions.[a]
À
into the Ni Ccage bonds in nickel–carborynes, and that the
final metal complex was a Ni0 species.[4–6] Also, 1-bromo-2-
lithiocarborane is a known precursor of carboryne.[2a–e]
Therefore, it is rational to assume that 1-bromo-2-lithiocar-
borane can undergo oxidative addition with Ni0 to give the
desired nickel–carboryne complex after elimination of LiBr.
Unfortunately, such an oxidative addition reaction does not
proceed at temperatures less than 08C and 1-bromo-2-
lithiocarborane is not stable at temperatures greater than
08C.[1,2a] Therefore, a new precursor to carboryne is required.
After many attempts, we discovered that 1-iodo-2-lithiocar-
borane is a good precursor for this catalytic cycle. Herein, we
report the nickel-catalyzed [2+2+2] cycloaddition of carbor-
yne with 2 equivalents of an alkyne to afford benzocarborane
compounds.
Entry Catalyst[b]
Loading [mol%] t [h] T [8C] Yield [%][c]
1
[Ni(cod)2]
20
20
20
20
20
20
10
20
20
20
20
20
20
20
20
20
2
2
2
2
2
2
2
4
4
2
2
2
2
2
2
2
110
110
110
110
110
110
110
110
90
110
110
110
110
110
110
110
49
33
37
17
57
65
31
63
60
29
22
16
1
2
3
[Ni(cod)2]/4PPh3
[Ni(PPh3)4]
4
5
6
7
8
9
10
11
12
13
14
15
16
[NiCl2(PMe3)2]
[NiCl2(PnBu3)]
[NiCl2(PPh3)2]
[NiCl2(PPh3)2]
[NiCl2(PPh3)2]
[NiCl2(PPh3)2]
[NiCl2(dppe)]
[NiCl2(dppp)]
[NiI2(Me2Im)2]
[Pd(PPh3)4]
[*] Z. Qiu, S. R. Wang, Prof. Dr. Z. Xie
Department of Chemistry and Center of Novel Functional
Molecules, The Chinese University of Hong Kong, Shatin
N.T., Hong Kong (China)
[PdCl2(PPh3)2]
[FeCl2]/2PPh3
[CoCl2(PPh3)2]
1
–
–
Fax: (+852)2603-5057
E-mail: zxie@cuhk.edu.hk
[**] This work was supported by grants from the Research Grants
Council of the Hong Kong Special Administration Region (Project
No. 404108), Direct Grant (Project No. 2060386), and The Chinese
University of Hong Kong.
[a] Conditions: 1) carborane (0.5 mmol), nBuLi (1.0 mmol), in toluene at
room temperature for 1 h. 2) I2 (0.5 mmol), at room temperature for
0.5 h; 3) catalyst, 3-hexyne (2 mmol). [b] cod=cyclooctadiene; dppe=
1,2-bis(diphenylphosphino)ethane. dppp=1,3-bis(diphenylphosphino)-
propane; Me2Im=1,3-dimethylimidazol-2-ylidene. [c] Yield of isolated
product.
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/anie.201001249.
Angew. Chem. Int. Ed. 2010, 49, 4649 –4652
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
4649

Communications
The NiII salts were also active, and their activities depended
observed for unsymmetrical arylalkynes 1g–l, presumably
owing to electronic effects as the phenyl group can be
considered as electron-withdrawing (Table 2, entries 9–14).[11]
When alkynes containing ether groups were employed in the
reaction (1e and 1m), the products were formed in low yields,
probably owing to the coordination of oxygen atoms occupy-
ing the vacant site on the nickel centre (Table 2, entries 7 and
15). Such interactions may also influence the regioselectivity
of the alkyne insertion and stabilize the inserted product,
which leads to the formation of 2’o and a small amount of
mono-alkyne insertion products after hydrolysis (Table 2,
entry 15).[12] Alkynes bearing an amido or carbonyl group,
such as 1n and 1o, were incompatible with this reaction
because they could react with the carboryne precursor 1-iodo-
2-lithiocarborane (Table 2, entries 16 and 17). For methyl 2-
butynoate, the homocyclotrimerization product was observ-
ed.[12a,b]
largely on the ligands around the nickel center (Table 1,
entries 4–12). [NiCl2(PPh3)2] was found to be the best catalyst,
producing 2a in 65% yield, thus suggesting that the Ni0
species that was generated in situ is more active than
[Ni(cod)2] (see below; Table 1, entry 6). Lower catalyst
loading (10 mol%) resulted in a significant decrease in the
yield of 2a from 65% to 31% (Table 1, entry 7). Prolonging
the reaction time from 2 to 4 hours did not affect the yield of
2a (Table 1, entry 8). Temperature was crucial to the reaction:
compound 2a was not observed at all if the reaction temper-
ature was below 608C. The reaction proceeded well at 908C,
but needed a longer time to proceed to completion (Table 1,
entry 9). In sharp contrast, palladium complexes, such as
[PdCl2(PPh3)2] and [Pd(PPh3)4], showed almost no activity
(Table 1, entries 13 and 14). [FeCl2]/PPh3 and [CoCl2(PPh3)2]
were inactive (Table 1, entries 15 and 16).
We then expanded the substrate scope to include various
carboranes and alkynes using the above optimum reaction
conditions (Table 1, entry 6), and the results are shown in
Table 2. The yields of 2 were comparable with those obtained
from the stoichiometric reactions of nickel–carboryne with
alkynes (Table 2, entries 1, 4–6, and 9).[4] Steric factors played
an important role in these reactions. Sterically less-demand-
ing 3-hexyne afforded the highest yield (Table 2, entry 1).
Carboranes with 3-chloro and 3-phenyl substituents showed a
big decrease in the yields of 2b,c from 65% to 31 and 38%,
respectively (Table 2, entries 2 and 3). 4-Methyl-2-pentyne 1 f
gave two inseparable regioisomers 2h/2’h in a molar ratio of
7:3 (Table 2, entry 8). However, excellent regioselectivity was
Internal diynes 3a–c were also compatible with these
nickel-catalyzed cycloaddition reactions and gave the desired
products 4 in 15–39% yields with a good tolerance of the
fused-ring size (Scheme 1). The yield was rather low for
seven-membered fused-ring species 4c, and no reaction
proceeded for the oxo-bridged diyne 3d.
Table 2: Nickel-catalyzed cycloaddition of carborynes with alkynes.
Scheme 1. Nickel-catalyzed cycloaddition of carboryne with diynes.
1
Compounds 2 and 4 were fully characterized by H, 13C,
and 11B NMR spectra, as well as high-resolution mass
spectrometry.[13] The molecular structures of 2h, 2n, and 4b
were further confirmed by single-crystal X-ray analyses (see
the Supporting Information).[14]
Entry
R1
R2
R3
1
Product Yield [%][a,b]
1
2
3
4
5
6
7
8
H
3-Cl
3-Ph
H
H
H
Et
Et
Et
nPr
nBu
Ph
Et
Et
Et
nPr
nBu
Ph
1a
1a
1a
1b
1c
1d
1e
2a
2b
2c
2d
2e
2 f
65 (67)
31
38
59 (65)
54 (65)
28 (33)
13
To gain some insight into the reaction mechanism, a
reaction of 1-I-2-Li-1,2-C2B10H10 with 1 equivalent of [Ni-
(cod)2]/4PPh3 was performed on an analytical scale in toluene
and monitored by 11B and 31P NMR spectroscopy. The results
suggested the formation of [(h2-C2B10H10)Ni(PPh3)2], even at
room temperature, which indicates the oxidative addition of
H
H
CH2OMe
iPr
CH2OMe
Me
2g
1 f 2h+2’h 44 (2h/2’h
=70:30)[c]
an I Ccage bond on the Ni0 center. Treatment of the in situ
À
9
H
H
H
H
H
H
H
H
H
Me
Me
Me
Et
Ph
1g
2i
2j
50 (54)
39
generated 1-I-2-Li-1,2-C2B10H10 with 1 equivalent of [NiCl2-
(PPh3)2] in the presence of 2 equivalents of n-butyl-2-pyridi-
nylacetylene in refluxing toluene gave mono-alkyne-insertion
10
11
12
13
14
15
16
17
p-Me-C6H4 1h
p-CF3-C6H4
1i
1j
1k
1l
2k
2l
49
49
Ph
Ph
Ph
Ph
Ph
Me
nBu
2m
2n
43
51
=
product 5 [{[2-C(nBu) C(o-C5H4N)-1,2-C2B10H10]Ni}2(m-Cl)]
ꢀ
C CPh
[Li(thf)4] after recrystallization from tetrahydrofuran as red
crystals in 25% yield (Scheme 2). This product was fully
characterized by various NMR spectroscopic techniques and
by elemental analysis.[15] Single-crystal X-ray analysis
revealed that 5 is an ionic complex that consists of dimeric
complex anions and tetrahedral cations. In the anion, two
square-planar nickel moieties share one m2-Cl atom
CH2OMe
CH2NMe2
CO2Me
1m 2o+2’o 24+2
1n
1o
–
–
–
–
[a] Yield of isolated product. [b] Yields in parentheses correspond to
those from the stoichiometric reactions of Ni-carboryne with 2 equiv-
alents of alkynes, reported in Ref. [4]. [c] Molar ratio was determined by
1H NMR analysis of the crude product mixture.
4650 www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 4649 –4652

Angewandte
Chemie
Scheme 2. Reaction with n-butyl-2-pyridinylacetylene.
(Figure 1). Coordination of the pyridinyl group to the nickel
atom can stabilize complex 5 and prevent the further insertion
of the second equivalent of n-butylpyridinylacetylene.
Scheme 3. Proposed mechanism for the nickel-catalyzed [2+2+2]
cyclization reaction.
In summary, we have developed the first metal-catalyzed
reaction of carboryne with unsaturated molecules using
1-iodo-2-lithiocarborane as a precursor and [NiCl2(PPh3)2]
as the catalyst. The mechanism was proposed after structural
confirmation of the key intermediate, nickelacyclopentene.
Experimental Section
Figure 1. Molecular structure of the anion in 5. Selected bond lengths
[ꢀ] and angles [8]: Ni1–C2 1.891(8), Ni1–C16 1.930 (8), Ni1–Cl1
2.267(2), Ni1–N2 1.966(7), C1–C2 1.656(12), C1–C11 1.487(11), C11–
C16 1.378(11), Ni2–C42 1.911(6), Ni2–C22 1.926(8), Ni2–Cl1 2.267(2),
Ni2–N1 1.946(6), C41–C42 1.641 (11), C41–C23 1.506 (10), C23–C22
1.345 (9); C2-Ni1-C16 86.8(3), C16-Ni1-Cl1 95.8(2), Cl1-Ni1-N2
83.8(2), N2-Ni1-C2 96.7(3), C42-Ni2-C22 85.6(3), C22-Ni2-Cl1 97.1(2),
Cl1-Ni2-N1 82.7(2), N1-Ni2-C42 97.6(3), Ni1-Cl1-Ni2 70.8(1).
Representative procedure: I2 (0.5 mmol) was added to a solution of
Li2C2B10H10 (0.5 mmol) in toluene (5 mL), prepared in situ from the
reaction of nBuLi (1.0 mmol) with ortho-carborane (0.5 mmol), and
the reaction mixture was stirred at room temperature for 0.5 h.
[NiCl2(PPh3)2] (0.1 mmol), and either the alkyne (2.0 mmol) or diyne
(1.0 mmol) were then added and the reaction vessel was closed and
heated at 1108C overnight. After addition of 5 mL of water and
extraction with ether (3 ꢀ 10 mL), the resulting solution was concen-
trated in vacuo. The residue was purified by column chromatography
on silica gel (230–400 mesh) using n-hexane as eluent to give the
cycloaddition product.
Given the above experimental evidence, a plausible
mechanism for the nickel-catalyzed cycloaddition is shown
in Scheme 3. The catalysis is likely to be initiated by a Ni0
species that is generated from the reduction of NiII with a
Received: March 2, 2010
Revised: April 3, 2010
Published online: May 12, 2010
lithiocarborane salt.[16] Oxidative addition between the I Ccage
À
Keywords: alkynes · carboryne · cycloaddition · nickel ·
.
bond and Ni0, followed by the subsequent elimination of
lithium iodide, produces nickel–carboryne intermediate B.
An alternative pathway proceeded through the elimination of
lithium iodide to form carboryne, and subsequent coordina-
tion to the metal center cannot be ruled out. Insertion of the
regioselectivity
[1] H. L. Gingrich, T. Ghosh, Q. Huang, M. Jones, Jr., J. Am. Chem.
Soc. 1990, 112, 4082 – 4083.
À
first equivalent of alkyne into the nickel–carboryne Ni Ccage
[2] a) T. Ghosh, H. L. Gingrich, C. K. Kam, E. C. M. Mobraaten, M.
Jones, Jr., J. Am. Chem. Soc. 1991, 113, 1313 – 1318; b) Q.
Huang, H. L. Gingrich, M. Jones, Jr., Inorg. Chem. 1991, 30,
3254 – 3257; c) R. J. Cunningham, N. Bian, M. Jones, Jr., Inorg.
Chem. 1994, 33, 4811 – 4812; d) D. M. Ho, R. J. Cunningham,
J. A. Brewer, N. Bian, M. Jones, Jr., Inorg. Chem. 1995, 34,
5274 – 5278; e) L. Barnett-Thamattoor, G. Zheng, D. M. Ho, M.
Jones, Jr., J. E. Jackson, Inorg. Chem. 1996, 35, 7311 – 7315; f) J.
Jeon, T. Kitamura, B.-W. Yoo, S. O. Kang, J. Ko, Chem.
bond gives nickelacyclopentene intermediate C. The second
À
equivalent of alkyne inserts into the Ni Cvinyl bond to afford
the seven-membered intermediate D.[4,17] Reductive elimina-
tion yields the cycloaddition product 2 and releases a Ni0
species to complete the catalytic cycle. The regioselectivity
observed in these reactions can be rationalized by the polarity
of alkynes.[11]
Angew. Chem. Int. Ed. 2010, 49, 4649 –4652
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
4651

Communications
Commun. 2001, 2110 – 2111; g) T. Lee, J. Jeon, K. H. Song, I.
Jung, C. Baik, K.-M. Park, S. S. Lee, S. O. Kang, J. Ko, Dalton
Trans. 2004, 933 – 937.
Hockless, E. Wenger, J. Chem. Soc. Dalton Trans. 1997, 3105 –
3114.
[12] For mono-insertion products, see the Supporting Information.
For the Ni0-catalyzed homocyclotrimerization of methyl 2-
butynoate, see: a) N. E. Schore, Chem. Rev. 1988, 88, 1081 –
1119; b) A. K. Jhingan, W. F. Maier, J. Org. Chem. 1987, 52,
1161 – 1165.
[3] A. A. Sayler, H. Beall, J. F. Sieckhaus, J. Am. Chem. Soc. 1973,
95, 5790 – 5792.
[4] L. Deng, H.-S. Chan, Z. Xie, J. Am. Chem. Soc. 2005, 127,
13774 – 13775.
[13] Experimental details and complete characterization data are
provided in the Supporting Information.
[5] Z. Qiu, Z. Xie, Angew. Chem. 2008, 120, 6674 – 6677; Angew.
Chem. Int. Ed. 2008, 47, 6572 – 6575.
[14] CCDC 767698 (2h), CCDC 767699 (2n), CCDC 767700 (4b),
and CCDC 767701 (5) contain the supplementary crystallo-
graphic 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.
[6] Z. Qiu, Z. Xie, J. Am. Chem. Soc. 2009, 131, 2084 – 2085.
[7] a) S. L. Buchwald, R. B. Nielsen, Chem. Rev. 1988, 88, 1047 –
1058; b) M. A. Bennett, H. P. Schwemlein, Angew. Chem. 1989,
101, 1349 – 1373; Angew. Chem. Int. Ed. Engl. 1989, 28, 1296 –
1320; c) M. A. Bennett, E. Wenger, Chem. Ber. 1997, 130, 1029 –
1042; d) W. M. Jones, J. Klosin, Adv. Organomet. Chem. 1998, 42,
147 – 221; e) M. Retbøll, A. J. Edwards, A. D. Rae, A. C. Willis,
M. A. Bennett, E. Wenger, J. Am. Chem. Soc. 2002, 124, 8348 –
8360.
[8] a) H. Wang, H.-W. Li, X. Huang, Z. Lin, Z. Xie, Angew. Chem.
2003, 115, 4483 – 4485; Angew. Chem. Int. Ed. 2003, 42, 4347 –
4349; b) L. Deng, H.-S. Chan, Z. Xie, J. Am. Chem. Soc. 2006,
128, 7728 – 7729; c) S. Ren, H.-S. Chan, Z. Xie, J. Am. Chem. Soc.
2009, 131, 3862 – 3863; d) S. Ren, L. Deng, H.-S. Chan, Z. Xie,
Organometallics 2009, 28, 5749 – 5756; e) S. Ren, H.-S. Chan, Z.
Xie, Organometallics 2009, 28, 4106 – 4114.
[9] Selected examples, see: a) D. Peꢁa, S. Escudero, D. Pꢂrez, E.
Guitian, L. Castedo, J. Am. Chem. Soc. 1999, 121, 5827 – 5828;
b) K. V. Radhakrishnan, E. Yoshikawa, Y. Yamamoto, Tetrahe-
dron Lett. 1999, 40, 7533 – 7535; c) E. Yoshikawa, K. V. Rad-
hakrishnan, Y. Yamamoto, J. Am. Chem. Soc. 2000, 122, 7280 –
7286; d) E. Yoshikawa, Y. Yamamoto, Angew. Chem. 2000, 112,
185 – 187; Angew. Chem. Int. Ed. 2000, 39, 173 – 175; e) D. Peꢁa,
D. Pꢂrez, E. Guitiꢃn, L. Castedo, J. Org. Chem. 2000, 65, 6944 –
6950; f) Y. Sato, T. Tamura, M. Mori, Angew. Chem. 2004, 116,
2490 – 2494; Angew. Chem. Int. Ed. 2004, 43, 2436 – 2440; g) Z.
Liu, X. Zhang, R. C. Larock, J. Am. Chem. Soc. 2005, 127,
15716 – 15717; h) J.-C. Hsieh, C.-H. Cheng, Chem. Commun.
2005, 2459 – 2461; i) J.-C. Hsieh, C.-H. Cheng, Chem. Commun.
2008, 2992 – 2994.
[15] a) Complete characterization data are provided in the Support-
ing
Information;
b) Crystal
data
for
5·THF:
ꢀ
C46H86B20ClLiN3Ni2O5, Mr = 1123.2, triclinic, space group P1,
a = 11.388(3), b = 14.951(3), c = 19.334(4) ꢄ, a = 86.53(1), b =
83.80(1), g = 74.03(1)8, V= 3145(1) ꢄ3, T= 296 K, Z = 2,
1calcd = 1.186 gcmÀ3, 2qmax = 508, m(MoKa) = 0.71073 ꢄ. A total
of 17044 reflections were collected and led to 11023 unique
reflections, 11023 of which with I > 2s(I) were considered as
observed, R1 = 0.0800, wR2 (F2) = 0.2052. This structure was
solved by direct methods and refined by full-matrix least-squares
on F2 by using the SHELXTL/PC package of crystallographic
software.[18] All non-hydrogen atoms were refined anisotropi-
cally and all hydrogen atoms were geometrically fixed using the
riding model.
[16] a) M. Miyaura in Topics in Current Chemistry: “Cross-Coupling
Reactions: A Practical Guide”, Springer, Berlin, 2002; b) J.
Terao, M. Tomita, S. P. Singh, N. Kambe, Angew. Chem. 2010,
122, 148 – 151; Angew. Chem. Int. Ed. 2010, 49, 144 – 147.
[17] The insertion of alkynes and other unsaturated molecules into
À
the M C(cage)bond in metal–carboranyl complexes is prohib-
ited because of steric reasons; see: a) Z. Xie, Coord. Chem. Rev.
2006, 250, 259 – 272; b) Y. Sun, H.-S. Chan, H. Zhao, Z. Lin, Z.
Xie, Angew. Chem. 2006, 118, 5659 – 5662; Angew. Chem. Int.
Ed. 2006, 45, 5533 – 5536; c) Y. Sun, H.-S. Chan, Z. Xie,
Organometallics 2006, 25, 3447 – 3453; d) Z. Qiu, Z. Xie, Sci.
China Ser. B 2009, 52, 1544 – 1558; e) H. Shen, Z. Xie, Chem.
Commun. 2009, 2431 – 2445.
[10] J. H. Atkins, D. M. Ho, M. Jones, Jr., Tetrahedron Lett. 1996, 37,
7217 – 7220.
[11] a) M. A. Bennett, S. A. Macgregor, E. Wenger, Helv. Chim. Acta
2001, 84, 3084 – 3104; b) M. A. Bennett, M. Glewis, D. C. R.
[18] G. M. Sheldrick, SHELXTL 5.10 for Windows NT: Structure
Determination Software Programs. Bruker Analytical X-ray
Systems, Inc., Madison, Wisconsin, USA, 1997.
4652
www.angewandte.org
ꢀ 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2010, 49, 4649 –4652