Isolation of Dinuclear (μ-Silylene)(silyl)nickel) Complexes and Si-Si Bond Formation on a Dinuclear Nickel Framework

Full text of DOI:10.1002/1521-3773(20010105)40:1<213::AID-ANIE213>3.0.CO;2-W

COMMUNICATIONS
qualitatively identified by using 1H NMR spectroscopy and GC-mass
spectrometry.
established.[2] The understanding of the structure and the
reactivity of silylnickel species[3] is indispensable for the
development of new catalytic methodologies. Although a
number of mononuclear silylmetal complexes are known for
all Group 10 metals, structurally characterized silylene-bridg-
ed dinuclear complexes have been reported only for plati-
num[4] and palladium.[5, 6] These dinuclear species are consid-
ered to play important roles in catalytic processes. For
example, silylene-bridged Pt2Si2 rings are potential key
intermediates in the platinum-catalyzed dehydrocoupling of
hydrosilanes.[4b±d] Here we report on the reaction of 1,2-
disilylbenzene (1) with nickel(0) complexes, which led to the
isolation of the dinuclear (m-silylene)(silyl)nickel(iii) com-
plexes [{1,2-C6H4(SiH2)(SiH)}2Ni2(R2PCH2CH2PR2)2] (R ˆ
Me or Et), which contain the first structurally characterized
four-membered Ni2Si2 rings. The short contact between the
SiH2 and m-SiH moieties revealed by X-ray crystallography
Received: July 31, 2000 [Z15561]
[1] a) NBC/NRC Steam Tables (Eds.: L. Haar, J. S. Gallagher, G. S. Kell),
Hemisphere, New York, 1984; b) High-Temperature Aqueous Solu-
tions: Thermodynamic Properties (Eds.: P. R. J. Fernandez, H. R.
Corti, M. L. Japer), CRC, Boca Raton, 1992; c) R. W. Shaw, T. B. Brill,
A. A. Clifford, C. A. Eckert, E. U. Franck, Chem. Eng. News 1991, 69
(51), 26.
[2] P. E. Savage, Chem. Rev. 1999, 99, 603.
[3] a) O. Sato, Y. Ikushima, T. Yokoyama, J. Org. Chem. 1998, 63, 9100;
b) Y. Ikushima, O. Sato, K. Hatakeda, T. Yokoyama, M. Arai, Angew.
Chem. 1999, 111, 3087; Angew. Chem. Int. Ed. 1999, 38, 2910; c) Y.
Ikushima, O. Sato, K. Hatakeda, T. Yokoyama, M. Arai, J. Am. Chem.
Soc. 2000, 122, 1908.
[4] D. Bröll, C. Kaul, A. Krämer, P. Krammer, T. Richter, M. Jung, H.
Vogel, P. Zehner, Angew. Chem. 1999, 111, 3180; Angew. Chem. Int.
Ed. 1999, 38, 2998.
[5] Y. Ikushima, K. Hatakeda, N. Saito, M. Arai, J. Chem. Phys. 1998, 108,
5855.
ꢀ
suggests the possibility of Si Si bond formation, which indeed
[6] a) S. Meyerson, P. N. Rylander, E. L. Eliel, J. D. McCollum, J. Am.
Chem. Soc. 1959, 81, 2606; b) E. L. Eliel, J. D. McCollum, S.
Meyerson, P. N. Rylander, J. Am. Chem. Soc. 1961, 83, 2481.
[7] a) C. G. Swain, A. L. Powell, T. J. Lynch, S. R. Alpha, R. P. Dunlap, J.
Am. Chem. Soc. 1979, 101, 3576, and refs therein; b) E. C. Ashby, D. T.
Coleman, M. Gamasa, Tetrahedron Lett. 1983, 24, 851; c) E. C. Ashby,
D. T. Coleman, M. Gamasa, J. Org. Chem. 1987, 52, 4079; d) S.-K.
Chung, J. Chem. Soc. Chem. Commun. 1982, 480; e) J. Weiss, Trans.
Faraday Soc. 1941, 37, 782; f) C. G. Swain, A. L. Powell, T. J. Lynch,
S. R. Alpha, R. P. Dunlap, J. Am. Chem. Soc. 1979, 101, 3584.
[8] C. R. Hauser, P. J. Hamrick, Jr., A. T. Stewart, J. Org. Chem. 1956, 21,
260.
takes place in a similar reaction of 1-(dimethylsilyl)-2-
silylbenzene (2) with [Ni(dmpe)2] (dmpe ˆ 1,2-bis(dimethyl-
phosphanyl)ethane).
[9] Y. Saeki, K. Negoro, S. Ichiyasu, Nippon Kagaku Kaishi 1985, 931.
[10] M. J. Kremer, K. A. Connery, C. M. DiPippo, J. Feng, J. E. Chateau-
neuf, J. F. Brennecke, J. Phys. Chem. A 1999, 103, 6591.
[11] a) N. Matsubayashi, C. Wakui, M. Nakahara, Phys. Rev. Lett. 1997, 78,
2573; b) M. M. Hoffmann, M. S. Conradi, J. Am. Chem. Soc. 1997, 119,
3811; c) H. Ohtaki, T. Radnai, T. Yamaguchi, Chem. Soc. Rev. 1997, 26,
41; d) Yu. E. Gorbaty, A. G. Kalinichev, J. Phys. Chem. 1995, 99, 5336.
Previously, we reported that treatment of
1 with
[Ni(dmpe)2] results in the dimeric bis(silyl)nickel(ii) complex
3 and the first tetrakis(silyl)nickel(iv) complex 4a.[7] A similar
reaction under somewhat different conditions led to the
formation of a new dimeric NiIII complex. While studying 3 by
variable-temperature NMR spectroscopy in [D8]toluene, we
observed, upon heating the solution to 808C, a color change
from pale yellow to orange and the emergence of new weak
Isolation of Dinuclear (m-Silylene)(silyl)nickel
1
signals in the H and 31P NMR spectra. Further heating at
ꢀ
Complexes and Si Si Bond Formation on a
1108C for 30 min gave a new complex 5a, which could be
isolated as orange crystals (34%) by crystallization from
benzene. The similar complex 5b, ligated by 1,2-bis(diethyl-
phosphanyl)ethane (depe), was also obtained directly from 1
and [Ni(PEt3)2(depe)] (1:Ni ˆ 1:1.05); the reaction proceeded
even at room temperature and gave 5b as the main product
(49% yield). The 29Si NMR spectrum of the crude mixture
showed the presence of the tetrakis(silyl)nickel(iv) complex
Dinuclear Nickel Framework**
Shigeru Shimada, Maddali L. N. Rao,
Teruyuki Hayashi, and Masato Tanaka*
Palladium and platinum complexes are versatile catalysts
for the transformation of organosilicon compounds.[1] How-
ever, the utility of nickel complexes has not been as well
4b as
a
side product [d ˆ ꢀ2.11 (dd, 2J(Si,Pcis) ˆ 20,
2
2J(Si,Ptrans) ˆ 101 Hz), 3.03 (t, J(Si,P) ˆ 16 Hz)].
[*] Prof. M. Tanaka, Dr. S. Shimada, Dr. M. L. N. Rao, Dr. T. Hayashi
National Institute of Materials and Chemical Research
Tsukuba, Ibaraki 305-8565 (Japan)
Fax : (‡ 81)298-61-4430
E-mail: mtanaka@home.nimc.go.jp
[**] We are grateful to the Japan Science and Technology Corporation
(JST) for financial support through the CREST (Core Research for
Evolutional Science and Technology) program and for a postdoctoral
fellowship to M.L.N.R.
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
Angew. Chem. Int. Ed. 2001, 40, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4001-0213 $ 17.50+.50/0
213

COMMUNICATIONS
The structures of 5a and 5b were established by X-ray
diffraction as dinuclear (m-silylene)(silyl)nickel complexes
(Figure 1).[8] In sharp contrast to the case of platinum, which
forms complex 6 with a mixed valent PtIIPtIVSi2 four-
membered ring,[4f] 5a and 5b have symmetric structures with
and a number of other signals were observed.[12] These results
suggest that 5b is in equilibrium between several species in
solution, although their structures are unknown at the mo-
ment.
The short Si1 ´´´ Si4 and Si2 ´´´ Si3 distances in 5a and 5b
ꢀ
suggest that and Si Si bond-forming reaction can possibly
take place in these complexes to form an eight-membered
cyclic dimer of 1. Attempted thermal
reactions of 5a or 5b did not provide
ꢀ
any clear evidence for Si Si bond
ꢀ
formation. However, Si Si bond for-
mation did take place upon treatment
of 2[13] with [Ni(dmpe)2] to generate the
dimeric complex 8. Heating a mixture
of 2 and [Ni(dmpe)2] (2:[Ni(dmpe)2] ˆ
2:1) in benzene at 55 ± 808C for 4 d
resulted in the spontaneous separation
of yellow crystals of 8 (21% yield) from
the solution. 1H, 31P, and 29Si NMR
spectroscopy on the solution part of the reaction mixture
revealed that a large amount of [Ni(dmpe)2] still remained,
while 2 was completely consumed to give a mixture of silicon-
containing compounds. Crystalline 8 is virtually insoluble in
benzene. The structure determined by X-ray diffraction
unequivocally verifies that the silyl ligand originated from
Figure 1. Molecular structure of 5a (30% probability thermal ellipsoids).
NiI2IISi2 four-membered rings. Most of the complexes contain-
ing four-membered Pt2Si2 rings so far reported have short
diagonal Si ´´´ Si distances (2.55 ± 2.88 Š) and long Pt ´´´ Pt
distances (3.66 ± 4.05 Š).[4b±f] In contrast, 5a and 5b have long
diagonal Si ´´´ Si distances (2.980(1) Š for 5a and 2.998(1) Š
for 5b) and short Ni ´´´ Ni distances (2.6658(7) Š for 5a and
2.7201(7) Š for 5b). The Ni ´´´ Ni distances are longer than the
[8]
ꢀ
Si Si bond formation through dehydrocoupling (Figure 2).
ꢀ
usual Ni Ni single bonds but still within the range of known
ꢀ
Ni Ni bond lengths; this suggests a bonding interaction.[9] The
Ni ± Si distances range from 2.210(1) to 2.304(1) Š for 5a and
from 2.2098(6) to 2.2976(6) Š for 5b, and 2.304(1) Š is the
ꢀ
longest Ni Si bond so far reported. In contrast to the diagonal
Si ´´´ Si distances, the Si1 ´´´ Si4 and Si2 ´´´ Si3 distances
(2.693(2) Š and 2.685(1) Š for 5a and 2.7049(9) for 5b) are
ꢀ
short and nearly the same as the value for the longest Si Si
single bond.[10] Complexes 5a and 5b are envisioned to be
formed by the dehydrogenative dimeriza-
tion of bis(silyl)nickel(ii) complexes 7a
and 7b, respectively. In the reaction of 1
with [Ni(dmpe)2], the sterically less de-
manding dmpe easily intercepts 7a to
form stable pentacoordinate species such
as 3, while in the reaction of 1 with
Figure 2. Molecular structure of 8 (30% probability thermal ellipsoids).
ꢀ
ꢀ
Selected bond lengths [Š] and angles [8]: Ni1 Si1 2.263(2), Ni1 Si3
ꢀ
ꢀ
ꢀ
ꢀ
2.257(2), Ni1 P1 2.160(2), Ni1 P2 2.182(2), Ni1 P3 2.236(2), Si1 Si2
ꢀ
2.357(2), Si3 Si4 2.361(2); Si1-Ni1-Si3 76.83(6), P1-Ni1-P2 86.81(8), P1-
Ni1-Si1 90.24(7), P2-Ni1-Si3 90.89(7), P1-Ni1-P3 103.53(8), Ni1-Si1-Si2
116.48(8), Ni1-Si1-C1 124.1(2), Si2-Si1-C1 102.3(2), Ni1-Si3-Si4 119.82(8),
Ni1-Si3-C2 121.5(2), Si4-Si3-C2 101.1(2).
[Ni(PEt3)2(depe)], the bulkier depe and PEt3 ligands prob-
ably prevent the formation of analogous pentacoordinate
nickel species and allow 7b to dimerize even at room
temperature.
Solid-state CP/MAS NMR spectra of 5b displayed 31P
signals at d ˆ 48.6 and 52.0, reasonable for the structure, and
29Si signals at d ˆ ꢀ48.5 and 79.9, which can be assigned to
SiH2 and m-SiH, respectively. A solution of 5b in [D8]THF at
ꢀ408C also displayed signals similar to those found in the
solid-state NMR spectrum; a pair of triplets at d ˆ 47.4 and
50.1 in the 31P{1H} NMR spectrum and two signals at d ˆ
ꢀ52.0 (ddd, J ˆ 6, 13, 55 Hz, SiH2) and 92.2 (quasi dt, J ˆ 6,
67 Hz, SiH) in the 29Si{1H} spectrum.[11] However, these
signals became broad and/or weak at higher temperatures,
Although further studies are necessary to clarify the reaction
mechanism, an assumption of the intermediacy of a dinuclear
species 9 similar to 5 appears to rationalize the formation of 8
1
4
2
3
ꢀ
ꢀ
(Scheme 1). Formation of Si Si and Si Si bonds on the
dimeric nickel framework in 9 and the release of a [Ni(dmpe)]
fragment generate a monomeric product 10, which readily
dimerizes by coordination of dmpe to form 8. The release of
Ni(dmpe) fragment is likely to be assisted by the simultaneous
ꢀ
attack of 2 to regenerate 11. The successful Si Si bond
formation observed for 2, but not for 1, is presumably
214
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4001-0214 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 1

COMMUNICATIONS
crystals precipitated and were collected by filtration, washed with benzene
(3 Â 1 mL), and dried under vacuum to give 8 (125 mg, 20.5%). M.p. 178 ±
1838C (decomp); IR (KBr): nÄ ˆ 3033, 2963, 2903, 1958, 1637, 1420, 1295,
1248, 1139, 933, 892, 830, 804, 746, 717, 657, 635, 440 cmꢀ1; elemental
analysis calcd for C50H92Ni2P6Si8: C 49.18, H 7.59; found: C 48.95, H 7.77.
Received: July 19, 2000 [Z15488]
[1] Recent reviews: H. Yamashita, M. Tanaka, Bull. Chem. Soc. Jpn. 1995,
68, 403 ± 419; K. A. Horn, Chem. Rev. 1995, 95, 1317 ± 1350; H. K.
Sharma, K. H. Pannell, Chem. Rev. 1995, 95, 1351 ± 1374; P. Braun-
stein, M. Knorr, J. Organomet. Chem. 1995, 500, 21 ± 38; M. Suginome,
Y. Ito, J. Chem. Soc. Dalton Trans. 1998, 1925 ± 1934; J.-Y. Corey, J.
Braddock-Wilking, Chem. Rev. 1999, 99, 175 ± 292.
[2] Examples of nickel-catalyzed reactions of organosilicon compounds:
Y. Kiso, K. Tamao, M. Kumada, J. Organomet. Chem. 1974, 76, 95 ±
103; K. Tamao, N. Miyake, Y. Kiso, M. Kumada, J. Am. Chem. Soc.
1975, 97, 5603 ± 5605; M. Ishikawa, S. Matsuzawa, T. Higuchi, S.
Kamitori, K. Hirotsu, Organometallics 1985, 4, 2040 ± 2046; M.
Ishikawa, S. Okazaki, A. Naka, H. Sakamoto, Organometallics 1992,
11, 4135 ± 4139; A. Naka, M. Hayashi, S. Okazaki, A. Kunai, M.
Ishikawa, Organometallics 1996, 15, 1101 ± 1105.
Scheme 1. A possible mechanism for the formation of 8.
associated with the low stability of organosilicon to transition
metal bonds relative to that of hydridosilicon to transition
metal bonds.[14] Further mechanistic studies and catalytic
dimerization of 1 and 2[15] are now underway.
[3] Recent examples of silylnickel and silylenenickel complexes: M. M.
Brezinski, J. Schneider, L. J. Radonovich, K. J. Klabunde, Inorg.
Chem. 1989, 28, 2414 ± 2419; M. Denk, R. K. Hayashi, R. West, J.
Chem. Soc. Chem. Commun. 1994, 33 ± 34; S. Nlate, E. Herdtweck,
R. A. Fischer, Angew. Chem. 1996, 108, 1957 ± 1959; Angew. Chem.
Int. Ed. Engl. 1996, 35, 1861 ± 1863; Y. Kang, J. Lee, Y. K. Kong, S. O.
Kang, J. Ko, Chem. Commun. 1998, 2343 ± 2344; B. Gehrhus, P. B.
Hitchcock, M. F. Lappert, H. Maciejewski, Organometallics 1998, 17,
5599 ± 5601.
[4] a) M. Auburn, M. Ciriano, J. A. K. Howard, M. Murray, N. J. Pugh,
J. L. Spencer, F. G. A. Stone, P. Woodward, J. Chem. Soc. Dalton
Trans. 1980, 659 ± 666; b) E. A. Zarate, C. A. Tessier-Youngs, W. J.
Youngs, J. Am. Chem. Soc. 1988, 110, 4068 ± 4070; c) E. A. Zarate,
C. A. Tessier-Youngs, W. J. Youngs, J. Chem. Soc. Chem. Commun.
1989, 577 ± 578; d) R. H. Heyn, T. D. Tilley, J. Am. Chem. Soc. 1992,
114, 1917 ± 1919; e) M. J. Michalczyk, C. A. Recatto, J. C. Calabrese,
M. J. Fink, J. Am. Chem. Soc. 1992, 114, 7955 ± 7957; f) S. Shimada, M.
Tanaka, K. Honda, J. Am. Chem. Soc. 1995, 117, 8289 ± 8290.
[5] M. Suginome, Y. Kato, N. Takeda, H. Oike, Y. Ito, Organometallics
1998, 17, 495 ± 497; Y.-J. Kim, S.-C. Lee, J.-I. Park, K. Osakada, J.-C.
Choi, T. Yamamoto, Organometallics 1998, 17, 4929 ± 4931.
Experimental Section
5a: A solution of 3 (160 mg, 0.19 mmol) in toluene (8 mL) was heated at
1108C for 30 min under N2. During heating, the color of the mixture turned
from yellow to red. After removal of volatile substances under vacuum
(10ꢀ4 Torr), the residue was dissolved in hot benzene (2 mL), from which
remaining 3 crystallized (10 mg). After collection of the crystals by
filtration, the filtrate was concentrated and the residue was dissolved in hot
benzene (1 mL). A few seed crystals of 5a were added, and the hot benzene
solution was slowly cooled to room temperature. During the cooling
process, orange prismatic crystals of 5a grew initially, and then a small
amount of crystalline 3 precipitated. After removal of the liquid portion
with a syringe, washing the solid with benzene (1 mL) and diethyl ether
(2 Â 2 mL), and drying under vacuum, the solids were manually separated
under a microscope to give pure 5a (50 mg, 34% yield) and a mixture of 3
and 5a (18 mg). 5a: M.p. 134 ± 1528C (decomp); IR (KBr): nÄ ˆ 3088, 3033,
2962, 2896, 2799, 2022, 1989, 1415, 1293, 1278, 1123, 1097, 931, 891, 819, 745,
701, 673, 632, 478, 459 cmꢀ1. X-ray diffraction showed the presence of
one molecule of benzene per molecule of 5a in the crystal; elemental
analysis calcd for C24H46Ni2P4Si4 ´ C6H6: C 47.02, H 6.84; found: C 46.78,
H 6.85.
[6] Although Lappert and Speier reported the formation of [(PPh3)2Ni(m-
SiCl2)2Ni(PPh3)2], only the elemental analyses were reported without
any spectral data to support the proposed structure: M. F. Lappert, G.
Speier, J. Organomet. Chem. 1974, 80, 329 ± 339.
[7] S. Shimada, M. N. L. Rao, M. Tanaka, Organometallics 1999, 18, 291 ±
293.
[8] Crystal structure analysis of 5a ´ C6H6: C24H46Ni2P4Si4 ´ C6H6, Mr ˆ
5b: [Ni(PEt3)4] (505 mg, 0.95 mmol) was dissolved in toluene (3 mL) and
depe (222 mL, 0.95 mmol) added at room temperature. After 20 min of
stirring, volatile substances were removed under vacuum to give [Ni-
(PEt3)2(depe)] (31P NMR ([D8]toluene, 202 MHz): d ˆ 16.25 (t, J ˆ 28 Hz),
30.98 (t, J ˆ 28 Hz)). 1 (124 mg, 0.90 mmol) was added to a solution of
crude [Ni(PEt3)2(depe)] in hexane (4 mL) at 08C. During the addition of 1,
the color of the mixture changed from light purple to light brown with
evolution of a small amount of H2. After the addition of 1, the mixture was
stirred at 308C for 44 h and then at 508C for 9 h to give a deep brown
solution. Removal of volatile substances under vacuum left a brown oil,
from which 5b crystallized. Washing the solid with diethyl ether (1 Â
15 mL, 2 Â 3 mL) followed by drying under vacuum afforded analytically
pure 5b as orange crystals (120 mg). Another crop (55 mg) of 5b was
obtained by evaporation of the diethyl ether washings and crystallization of
the residue from diethyl ether/hexane. Total yield based on 1 was 49%.
M.p. 150 ± 1578C (decomp); 31P CP/MAS NMR (121.5 MHz): d ˆ 48.6,
52.0; 29Si CP/MAS NMR (59.6 MHz): d ˆ ꢀ48.5, 79.8; IR (KBr): nÄ ˆ 3034,
2961, 2932, 2902, 2028, 1456, 1416, 1376, 1240, 1097, 1026, 950, 867, 816, 762,
677, 622 cmꢀ1; elemental analysis calcd for C32H62Ni2P4Si4: C 48.02, H 7.81;
found: C 48.33, H 7.75.
766.38, monoclinic, space group P21/c, a ˆ 11.412(3), b ˆ 16.510(3),
c ˆ 20.489(4) Š, b ˆ 91.03(2)8, Vˆ 3859(1) Š3, Z ˆ 4, 1calcd
ˆ
1.32 gcmꢀ3, R ˆ 0.038, Rw ˆ 0.047. Crystal structure analysis of 5b:
C32H62Ni2P4Si4, Mr ˆ 800.48, monoclinic, space group C2/c, a ˆ
20.222(3), b ˆ 10.974(4), c ˆ 19.785(4) Š, b ˆ 115.45(1)8, Vˆ
3964(1) Š3, Z ˆ 4, 1calcd ˆ 1.34 gcmꢀ3, R ˆ 0.029, Rw ˆ 0.039. Crystal
structure analysis of 8: C50H92Ni2P6Si8, Mr ˆ 1221.20, triclinic, space
Å
group P1, a ˆ 9.488(3), b ˆ 11.327(3), c ˆ 16.766(3) Š, a ˆ 88.92(2),
b ˆ 81.06(2), g ˆ 68.25(2)8, Vˆ 1651.7(9) Š3, Z ˆ 1, 1calcd ˆ 1.23 gcmꢀ3
,
R ˆ 0.046, Rw ˆ 0.054. Crystallographic data (excluding structure
factors) for the structures reported in this paper have been deposited
with the Cambridge Crystallographic Data Centre as supplementary
publications nos. CCDC-138595 (5a), -138596 (5b) and -138597 (8).
Copies of the data can be obtained free of charge on application to
CCDC, 12 Union Road, Cambridge CB21EZ, UK (fax: (‡44)1223-
336-033; e-mail: deposit@ccdc.cam.ac.uk).
[9] G. Henkel, M. Kriege, K. Matsumoto, J. Chem. Soc. Dalton Trans.
1988, 657 ± 659; T. A. Wark, D. W. Stephan, Organometallics 1989, 8,
2836 ± 2843; T. Saito, Y. Kajitani, T. Yamagata, N. Imoto, Inorg. Chem.
1990, 29, 2951 ± 2955.
8: A solution of 2 (0.332 g, 2 mmol) and [Ni(dmpe)2] (0.358 g, 1 mmol) in
benzene (4 mL) was heated at 558C for 24 h and at 808C for 3 d. Yellow
[10] N. Wiberg, H. Schuster, A. Simon, K. Peters, Angew. Chem. 1986, 98,
100 ± 101; Angew. Chem. Int. Ed. Engl. 1986, 25, 79 ± 80.
Angew. Chem. Int. Ed. 2001, 40, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4001-0215 $ 17.50+.50/0
215

COMMUNICATIONS
1
[11] See the Supporting Information for the variable-temperature H, 31P,
and 29Si NMR spectra of 5b.
[12] 1H and 31P NMR spectra of 5a in C6D6 were also complex at room
temperature.
[13] Preparation of compound 2 was based on Tamaoꢁs method: K. Tamao,
H. Yao, Y. Tsutsumi, H. Abe, T. Hayashi, Y. Ito, Tetrahedron Lett.
1990, 31, 2925 ± 2928.
ꢀ
ꢀ
[14] It is known that M SiH3 bonds are shorter than M SiMe3 bonds (M ˆ
(CO)5Mn or (CO)5Re): L. Manojlovic-Muir, K. W. Muir, J. A. Ibers,
Inorg. Chem. 1970, 9, 447 ± 452; D. W. H. Rankin, A. Robertson, J.
Organomet. Chem. 1975, 85, 225 ± 235; D. W. H. Rankin, A. Robert-
son, J. Organomet. Chem. 1976, 105, 331 ± 340; M. C. Couldwell, J.
Simpson, W. T. Robinson, J. Organomet. Chem. 1976, 107, 323 ± 339.
[15] A preliminary study showed that nickel complexes catalyze the
dehydrogenative dimerization of 2.
A Mass Spectrometric Labeling Strategy
for High-Throughput Reaction Evaluation
and Optimization:
ꢀ
Exploring C H Activation**
ꢀ
Scheme 2. a) A mixture of diverse pyridines is C H activated to provide a
mixture of products each bearing ketone group. b) A schematic
Jason W. Szewczyk, Rebecca L. Zuckerman,
Robert G. Bergman,* and Jonathan A. Ellman*
a
representation summarizing the high-throughput strategy for the optimi-
zation of the reaction and discovery of new products.
[1]
ꢀ
Chemical transformations that combine C H activation
nitrogen atom coordinates the ruthenium catalyst, and directs
and olefin insertion[2] present attractive opportunities for the
rapid assembly of complex structures from unexploited
classes of building blocks through carbon ± carbon bond
formation (Scheme 1).[3] Controlling the site of activation is
activation specifically to the ortho sp2 C H bonds. More
ꢀ
recently, Murai and co-workers have extended this reaction to
several different aromatic heterocycles and olefin sub-
strates.[5] A better understanding of the rules for functional
group compatibility and substrate generality demands that a
large number of heterocyclic classes be investigated, yet
current approaches cannot efficiently provide this informa-
tion because of the large number of compounds that require
testing.
M
R2
R1
H
+
R2
R1
O
M, CO
R1
H
R1
R2
Combinatorial methods have made a major impact on new
catalyst discovery, but the development of rapid and efficient
analyses remains a challenge.[7] Several recent high-through-
put methods monitor reactions through the catalyst turnover
and/or reaction conversion by IR thermography,[8] the for-
mation of UV-active products,[9] or the production of acid[10]
or carbon dioxide.[11] Although, these techniques are power-
ful, they cannot distinguish between products, by-products,
and decomposition. Alternatively, serial methods (TLC, GC,
and HPLC) have been employed to identify products and/or
quantify yields,[12] but these methods severely limit the
number of reactions that can be analyzed in a short period
of time. In addition, fluorescent-labeled substrates have been
used in the rapid monitoring of reactions.[13] Unfortunately,
each substrate under investigation must be individually
synthesized as a specifically labeled compound prior to
reaction and analysis; thus experimental effort increases
linearly with the number of different substrates to be
evaluated. We report here the first optimization strategy that
enables efficient quantitation of product yields at multiple
time points for large numbers of substrates, thereby establish-
ing structure ± reactivity relationships and reaction compati-
bility of functional groups (Scheme 2b).
R2
+
ꢀ
Scheme 1. The generality of C H activation allows, in principle, the
combination of aromatic heterocycles with alkenes to rapidly assemble
diverse structures using unexploited classes of building blocks.
one of the key challenges of this chemistry. Moore et al. first
demonstrated that [Ru3(CO)12] catalytically activates the
ortho positions of pyridine toward acylation with CO and
olefins (Scheme 2a).[4] They proposed that the aromatic
[*] Prof. R. G. Bergman, Prof. J. A. Ellman, J. W. Szewczyk,
R. L. Zuckerman
Department of Chemistry
University of California-Berkeley
Berkeley, CA 94720 (USA)
Fax: (‡1)510-642-8369
E-mail: bergman@cchem.berkeley.edu
jellman@uclink4.berkeley.edu
[**] This work was supported by the Director, Office of Energy Research,
Office of Basic Energy Sciences, Chemical Sciences Division, U.S.
Department of Energy, under contract No. DE-AC03-76SF00098. We
are grateful to the National Institutes of Health for support (GM-
50353) and for a Postdoctoral Fellowship to J.W.S.
Supporting information for this article is available on the WWW under
http://www.angewandte.com or from the author.
216
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
1433-7851/01/4001-0216 $ 17.50+.50/0
Angew. Chem. Int. Ed. 2001, 40, No. 1