A cheap metal for a "noble" task: Preparative and mechanistic aspects of cycloisomerization and cycloaddition reactions catalyzed by low-valent iron complexes

Article abstract of DOI:10.1021/ja0777180

Reaction of ferrocene with lithium in the presence of either ethylene or COD allows the Fe(0)-ate complexes 1 and 4 to be prepared on a large scale, which turned out to be excellent catalysts for a variety of Alder-ene, [4+2], [5+2], and [2+2+2] cycloaddi

Full text of DOI:10.1021/ja0777180

Published on Web 01/16/2008
A Cheap Metal for a “Noble” Task: Preparative and
Mechanistic Aspects of Cycloisomerization and Cycloaddition
Reactions Catalyzed by Low-Valent Iron Complexes
Alois Fu¨rstner,* Keisuke Majima, Rube´n Mart´ın, Helga Krause, Egmont Kattnig,
Richard Goddard, and Christian W. Lehmann
Max-Planck-Institut fu¨r Kohlenforschung, D-45470 Mu¨lheim/Ruhr, Germany
Received October 8, 2007; E-mail: fuerstner@mpi-muelheim.mpg.de
Abstract: Reaction of ferrocene with lithium in the presence of either ethylene or COD allows the Fe(0)-
ate complexes 1 and 4 to be prepared on a large scale, which turned out to be excellent catalysts for a
variety of Alder-ene, [4+2], [5+2], and [2+2+2] cycloadditon and cycloisomerization reactions of
polyunsaturated substrates. The structures of ferrates 1 and 4 in the solid-state reveal the capacity of the
reduced iron center to share electron density with the ligand sphere. This feature, coupled with the kinetic
lability of the bound olefins, is thought to be responsible for the ease with which different enyne or diyne
substrates undergo oxidative cyclization as the triggering event of the observed skeletal reorganizations.
This mechanistic proposal is corroborated by highly indicative deuterium labeling experiments. Moreover,
it was possible to intercept two different products of an oxidative cyclization manifold with the aid of the
Fe(+1) complex 6, which, despite its 17-electron count, also turned out to be catalytically competent in
certain cases. The unusual cyclobutadiene complex 38 derived from 6 and tolane was characterized by
X-ray crystallography.
ancillary ligands.4 More recently, however, a rapidly growing
number of examples highlights how iron catalysis may fertilize
Introduction
Late transition metals, and in particular the noble ones among
them, dominate a very significant part of contemporary catalysis
research. Their ability to maintain reversible redox cycles, the
excellent tunability of their steric and electronic characteristics
by a host of ligand sets, the pronounced affinity to π-systems,
and an outstanding compatibility with many functional groups
are just the most elementary of the many factors that contribute
to this privileged status.1,2 Yet, noble metals in general are (very)
expensive, whereas other commonly used late transition metals
are either toxic and/or require special treatment of the waste
stream due to serious environmental concerns (cobalt, nickel,
copper). Therefore, it is a worthwhile endeavor to search for
possible alternatives in a quest for more affordable and
sustainable methodology.
Iron catalysis seems to provide many opportunities in this
regard, because salts of this metal are cheap, generally nontoxic,
benign, and readily available. Despite these a priori favorable
attributes and the pre-eminent role of iron in biological
catalysis,3 applications in organic synthesis were largely con-
fined, for a long time, to Lewis acid chemistry as well as to the
stoichiometric use of iron templates stabilized by strongly bound
fields not commonly associated with this particular metal,5 such
as olefin oligo- and polymerization,6 cross coupling chemistry,7-16
(4) (a) Semmelhack, M. F. In Organometallics in Synthesis. A Manual, 2nd
ed.; Schlosser, M., Ed.; Wiley: Chichester, 2002; pp 1003. (b) Kno¨lker,
H.-J. Chem. ReV. 2000, 100, 2941. (c) Pearson, A. J. Iron Compounds in
Organic Synthesis; Academic Press: London, 1994. (d) Cox, L. R.; Ley,
S. V. Chem. Soc. ReV. 1998, 27, 301. (e) Ru¨ck-Braun, K.; Mikula´s, M.;
Amrhein, P. Synthesis 1999, 727.
(5) For a general review on applications of iron compounds in catalysis, see:
Bolm, C.; Legros, J.; Le Paih, J.; Zani, L. Chem. ReV. 2004, 104, 6217.
(6) (a) Small, B. L.; Brookhart, M. Macromolecules 1999, 32, 2120. (b) Small,
B. L.; Brookhart, M.; Bennett, A. M. A. J. Am. Chem. Soc. 1998, 120,
4049. (c) Britovsek, G. J. P.; Mastroianni, S.; Solan, G. A.; Baugh, S. P.
D.; Redshaw, C.; Gibson,V. C.; White, A. J. P.; Williams, D. J.; Elsegood,
M. R. J. Chem. Eur. J. 2000, 6, 2221. (d) Small, B. L.; Marcucci, A. J.
Organometallics 2001, 20, 5738. (e) Wakioka, M.; Baek, K.-Y.; Ando,
T.; Kamigaito, M.; Sawamoto, M. Macromolecules 2002, 35, 330. (f)
Bouwkamp, M. W.; Lobkovsky, E.; Chirik, P. J. J. Am. Chem. Soc. 2005,
127, 9660 and literature cited therein.
(7) Review: Fu¨rstner, A.; Martin, R. Chem. Lett. 2005, 34, 624.
(8) (a) Tamura, M.; Kochi, J. K. J. Am. Chem. Soc. 1971, 93, 1487. (b) Tamura,
M.; Kochi, J. Synthesis 1971, 303. (c) Neumann, S. M.; Kochi, J. K. J.
Org. Chem. 1975, 40, 599. (d) Smith, R. S.; Kochi, J. K. J. Org. Chem.
1976, 41, 502. (e) Kochi, J. K. Acc. Chem. Res. 1974, 7, 351.
(9) (a) Cahiez, G.; Avedissian, H. Synthesis 1998, 1199. (b) Dohle, W.; Kopp,
F.; Cahiez, G.; Knochel, P. Synlett 2001, 1901. (c) Cahiez, G.; Habiak, V.;
Duplais, C.; Moyeux, A. Angew. Chem., Int. Ed. 2007, 46, 4364. (d)
Duplais, C.; Bures, F.; Sapountzis, I.; Korn, T. J.; Cahiez, G.; Knochel, P.
Angew. Chem., Int. Ed. 2004, 43, 2968. (e) Cahiez, G.; Duplais, C.; Moyeux,
A. Org. Lett. 2007, 9, 3253.
(1) (a) Tsuji, J. Transition Metal Reagents and Catalysts. InnoVations in
Organic Synthesis; Wiley: Chichester, 2000. (b) Collman, J. P.; Hegedus,
L. S.; Norton, J. R.; Finke, R. G. Principles and Applications of
Organotransition Metal Chemistry, 2nd ed.; University Science Books: Mill
Valley, 1987. (c) Fundamentals of Molecular Catalysis; Kurosawa, H.,
Yamamoto, A., Eds.; Curr. Methods Inorg. Chem.; Elsevier: Amsterdam,
2003; Vol. 3.
(2) Fu¨rstner, A.; Davies, P. W. Angew. Chem., Int. Ed. 2007, 46, 3410.
(3) Bertini, I.; Gray, H. B.; Lippard, S. J.; Valentine, J. S. Bioinorganic
Chemistry; University Science Books: Mill Valley, 1994.
(10) Fu¨rstner, A.; Brunner, H. Tetrahedron Lett. 1996, 37, 7009.
(11) (a) Fu¨rstner, A.; Leitner, A.; Me´ndez, M.; Krause, H. J. Am. Chem. Soc.
2002, 124, 13856. (b) Fu¨rstner, A.; Leitner, A. Angew. Chem., Int. Ed.
2002, 41, 609. (c) Scheiper, B.; Bonnekessel, M.; Krause, H.; Fu¨rstner, A.
J. Org. Chem. 2004, 69, 3943. (d) Seidel, G.; Laurich, D.; Fu¨rstner, A. J.
Org. Chem. 2004, 69, 3950. (e) Fu¨rstner, A.; Me´ndez, M. Angew. Chem.,
Int. Ed. 2003, 42, 5355. (f) Fu¨rstner, A.; Schlecker, A.; Lehmann, C. W.
Chem. Commun. 2007, 4277.
(12) (a) Shinokubo, H.; Oshima, K. Eur. J. Org. Chem. 2004, 2081. (b) Hayashi,
Y.; Shinokubo, H.; Oshima, K. Tetrahedron Lett. 1998, 39, 63.
9
1992
J. AM. CHEM. SOC. 2008, 130, 1992-2004
10.1021/ja0777180 CCC: $40.75 © 2008 American Chemical Society

Cheap Metal for a “Noble” Task
A R T I C L E S
allylations,17 carbometalations,18 cycloadditions,19,20 cycloiso-
merizations,21 or hydrogenations,22 to name just a few.23 Despite
these important advances, the field is still in its infancy, not
least because the exact nature of the catalytically competent
iron species is often unknown and/or subject to somewhat
controversial debate.
Convinced of the relevance of this emerging field, we
launched a program aiming at a more systematic investigation
of the potential of iron catalysis for preparative purposes. So
far, this venture has successfully borne out in the cross coupling
arena,7,10,11,14 including various applications to the total synthesis
of bioactive natural products.24,25 Though highly efficacious in
many cases, such C-C bond formations are mechanistically
exceedingly complex and may occur along more than one
pathway.26 In an attempt to gain a better mechanistic under-
standing, we became interested in structurally well-defined low-
valent iron complexes as potential mimics for the actual catalysts
generated in situ from FeX3 and RMgX. In fact, the rather
unusual but structurally well-defined olefin complexes 1 and 2
turned out to be highly relevant in this context, because they
even allow one to effect the particularly difficult cross coupling
of alkyl halides and Grignard reagents with exceptional ease.27,28
Encouraged by these results, we started to investigate the yet
largely unexplored chemical behavior of such “bare” ferrate
species in more detail, which led to the discovery of their
remarkable potential as catalysts for various cycloisomerization
and cycloaddition reactions of polyunsaturated substrates.
Outlined below is a summary of our investigations in this field.29
Results and Discussion
(13) (a) Molander, G. A.; Rahn, B. J.; Shubert, D. C.; Bonde, S. E. Tetrahedron
Lett. 1983, 24, 5449. (b) Fakhfakh, M. A.; Franck, X.; Hocquemiller, R.;
Figade`re, B. J. Organomet. Chem. 2001, 624, 131. (c) Fabre, J.-L.; Julia,
M.; Verpeaux, J.-N. Tetrahedron Lett. 1982, 23, 2469. (d) Fiandanese, V.;
Marchese, G.; Martina, V.; Ronzini, L. Tetrahedron Lett. 1984, 25, 4805.
(e) Yanagisawa, A.; Nomura, N.; Yamamoto, H. Tetrahedron 1994, 50,
6017. (f) Quintin, J.; Franck, X.; Hocquemiller, R.; Figade`re, B. Tetrahedron
Lett. 2002, 43, 3547. (g) Nee`as, D.; Drabina, P.; Sedla´k, M.; Kotora, M.
Tetrahedron Lett. 2007, 48, 4539. (h) Ottesen, L. K.; Ek, F.; Olsson, R.
Org. Lett. 2006, 8, 1771. (i) Itami, K.; Higashi, S.; Mineno, M.; Yoshida,
J. Org. Lett. 2005, 7, 1219. (j) Østergaard, N.; Pedersen, B. T.; Skjærbæk,
N.; Vedsø, P.; Begtrup, M. Synlett 2002, 1889. (k) Pridgen, L. N.; Snyder,
L.; Prol, J., Jr. J. Org. Chem. 1989, 54, 1523. (l) Dos Santos, M.; Franck,
X.; Hocquemiller, R.; Figade`re, B.; Peyrat, J.-F.; Provot, O.; Brion, J.-D.;
Alami, M. Synlett 2004, 2697. (m) Ho¨lzer, B.; Hoffmann, R. W. Chem.
Commun. 2003, 732. (n) Gue´rinot, A.; Reymond, S.; Cossy, J. Angew.
Chem., Int. Ed. 2007, 46, 6521. (o) Dongol, K. G.; Koh, H.; Sau, M.; Chai,
C. L. L. AdV. Synth. Catal. 2007, 349, 1015.
(14) Fu¨rstner, A.; Leitner, A.; Seidel, G. Org. Synth. 2005, 81, 33.
(15) (a) Kofink, C. C., Blank, B.; Pagano, S.; Go¨tz, N.; Knochel, P. Chem.
Commun. 2007, 1954. (b) Dunet, G.; Knochel, P. Synlett 2006, 407. (c)
Sapountzis, I.; Lin, W.; Kofink, C. C.; Despotopoulos, C.; Knochel, P.
Angew. Chem., Int. Ed. 2005, 44, 1654.
(16) Hatakeyama, T.; Nakamura, M. J. Am. Chem. Soc. 2007, 129, 9844.
(17) (a) Plietker, B. Angew. Chem., Int. Ed. 2006, 45, 6053. (b) Plietker, B.
Angew. Chem., Int. Ed. 2006, 45, 1469. (c) Nakamura, M.; Matsuo, K.;
Inoue, T.; Nakamura, E. Org. Lett. 2003, 5, 1373. (d) Durandetti, M.;
Pe´richon, J. Tetrahedron Lett. 2006, 47, 6255.
(18) (a) Hojo, M.; Murakami, Y.; Aihara, H.; Sakuragi, R.; Baba, Y.; Hosomi,
A. Angew. Chem., Int. Ed. 2001, 40, 621. (b) Nakamura, M.; Hirai, A.;
Nakamura, E. J. Am. Chem. Soc. 2000, 122, 978. (c) Yamagami, T.;
Shintani, R.; Shirakawa, E.; Hayashi, T. Org. Lett. 2007, 9, 1045. (d)
Shirakawa, E.; Yamagami, T.; Kimura, T.; Yamaguchi, S.; Hayashi, T. J.
Am. Chem. Soc. 2005, 127, 17164. (e) Zhang, D.; Ready, J. M. J. Am.
Chem. Soc. 2006, 128, 15050.
(19) Pioneering studies on [4+2] cycloadditions: (a) Carbonaro, A.; Greco, A.;
Dall’Asta, G. Tetrahedron Lett. 1967, 8, 2037. (b) Carbonaro, A.; Greco,
A.; Dall’Asta, G. J. Org. Chem. 1968, 33, 3948. (c) Genet, J. P.; Ficini, J.
Tetrahedron Lett. 1979, 20, 1499. (d) tom Dieck, H.; Diercks, R. Angew.
Chem., Int. Ed. Engl. 1983, 22, 778.
Low-Valent Ferrate Complexes. Under the premise that iron
complexes of suitable oxidation state, spin configuration, and
coordination geometry might qualify as substitutes for precious
noble metal catalysts, we became interested in the pioneering
work of Jonas et al. who showed that the individual Cp-rings
of ferrocene can be successively removed under reducing
conditions.30,31 When performed under ethylene atmosphere, the
reductive cleavage primarily results in formation of the Fe(0)
half-sandwich complex 1, which reacts further on prolonged
stirring at ambient temperature to give the rather unusual
tetraethylene ferrate 2 of the formal oxidation state -2 (Scheme
1). Both compounds can be isolated in respectable yields on a
multigram scale as air-sensitive but nicely crystalline and
indefinitely storable materials upon complexation of the escort-
ing lithium counterions with TMEDA. If ethylene is replaced
by 1,5-cyclooctadiene (COD), complex 3 or the TMEDA-free
(24) (a) Fu¨rstner, A.; Leitner, A. Angew. Chem., Int. Ed. 2003, 42, 308. (b)
Scheiper, B.; Glorius, F.; Leitner, A.; Fu¨rstner, A. Proc. Natl. Acad. Sci.
U.S.A. 2004, 101, 11960. (c) Fu¨rstner, A.; De Souza, D.; Parra-Rapado,
L.; Jensen, J. T. Angew. Chem., Int. Ed. 2003, 42, 5358. (d) Lepage, O.;
Kattnig, E.; Fu¨rstner, A. J. Am. Chem. Soc. 2004, 126, 15970. (e) Fu¨rstner,
A.; Turet, L. Angew. Chem., Int. Ed. 2005, 44, 3462. (f) Fu¨rstner, A.;
Kattnig, E.; Lepage, O. J. Am. Chem. Soc. 2006, 128, 9194. (g) Fu¨rstner,
A.; De Souza, D.; Turet, L.; Fenster, M. D. B.; Parra-Rapado, L.; Wirtz,
C.; Mynott, R.; Lehmann, C. W. Chem.-Eur. J. 2007, 13, 115. (h) Fu¨rstner,
A.; Kirk, D.; Fenster, M. D. B.; A¨ıssa, C.; De Souza, D.; Nevado, C.; Tuttle,
C. T. T.; Thiel, W.; Mu¨ller, O. Chem.-Eur. J. 2007, 13, 135. (i) Fu¨rstner,
A.; Kirk, D.; Fenster, M. D. B.; A¨ıssa, C.; De Souza, D.; Mu¨ller, O. Proc.
Natl. Acad. Sci. U.S.A. 2005, 102, 8103. (j) Fu¨rstner, A.; Hannen, P.
Chem.-Eur. J. 2006, 12, 3006.
(20) (a): [2+2] cycloadditions: Bouwkamp, M. W.; Bowman, A. C.; Lobkov-
sky, E.; Chirik, P. J. J. Am. Chem. Soc. 2006, 128, 13340. (b) [4+4]
cycloadditions: Baldenius, K.-U.; tom Dieck, H.; Ko¨nig, W. A.; Icheln,
D.; Runge, T. Angew. Chem., Int. Ed. Engl. 1992, 31, 305. (c) Mallien,
M.; Haupt, E. T. K.; tom Dieck, H. Angew. Chem., Int. Ed. Engl. 1988,
27, 1062. (d) [4+1] cycloadditions: Sigman, M. S.; Kerr, C. E.; Eaton, B.
E. J. Am. Chem. Soc. 1993, 115, 7545. (e) Sigman, M. S.; Eaton, B. E. J.
Am. Chem. Soc. 1996, 118, 11783. (f) [4+2+2] cycloadditions: Greco,
A.; Carbonaro, A.; Dall’Asta, G. J. Org. Chem. 1970, 35, 271.
(21) For ene-type reactions of ene/diene substrates using iron catalysts formed
in situ see: (a) Takacs, J. M.; Newsome, P. W.; Kuehn, C.; Takusagawa,
F. Tetrahedron 1990, 46, 5507. (b) Takacs, J. M.; Anderson, L. G. J. Am.
Chem. Soc. 1987, 109, 2200. (c) Takacs, J. M.; Weidner, J. J.; Takacs, B.
E. Tetrahedron Lett. 1993, 34, 6219. (d) Takacs, J. M.; Weidner, J. J.;
Newsome, P. W.; Takacs, B. E.; Chidambaram, R.; Shoemaker, R. J. Org.
Chem. 1995, 60, 3473. (e) Takacs, J. M.; Myoung, Y.-C.; Anderson, L. G.
J. Org. Chem. 1994, 59, 6928. (f) Takacs, J. M.; Anderson, L. G.; Newsome,
P. W. J. Am. Chem. Soc. 1987, 109, 2542. (g) For an example of a
stoichiometric ene-reaction of enynes induced by iron carbonyl complexes,
see: Pearson, A. J.; Dubbert, R. A. Organometallics 1994, 13, 1656.
(22) For a leading reference see: Casey, C. P.; Guan, H. J. Am. Chem. Soc.
2007, 129, 5816.
(25) For other applications in synthesis, see: (a) Hamajima, A.; Isobe, M. Org.
Lett. 2006, 8, 1205. (b) Scheerer, J. R.; Lawrence, J. F.; Wang, G. C.;
Evans, D. A. J. Am. Chem. Soc. 2007, 129, 8968. (c) Dickschat, J. S.;
Reichenbach, H.; Wagner-Do¨bler, I.; Schulz, S. Eur. J. Org. Chem. 2005,
4141. (d) Hocek, M.; Dvora´kova´, H. J. Org. Chem. 2003, 68, 5773. (e)
Maulide, N.; Vanherck, J.-C.; Marko´, I. E. Eur. J. Org. Chem. 2004, 3962.
(26) Fu¨rstner, A.; Krause, H.; Lehmann, C. W. Angew. Chem., Int. Ed. 2006,
45, 440.
(27) Martin, R.; Fu¨rstner, A. Angew. Chem., Int. Ed. 2004, 43, 3955.
(28) For cross coupling reactions of alkyl halides catalyzed by structurally
unknown iron catalysts generated in situ, see: (a) Nakamura, M.; Matsuo,
K.; Ito, S.; Nakamura, E. J. Am. Chem. Soc. 2004, 126, 3686. (b) Nagano,
T.; Hayashi, T. Org. Lett. 2004, 6, 1297. (c) Nakamura, M.; Ito, S.; Matsuo,
K.; Nakamura, E. Synlett 2005, 1794. (d) Bedford, R. B.; Betham, M.;
Bruce, D. W.; Danopoulos, A. A.; Frost, R. M.; Hird, M. J. Org. Chem.
2006, 71, 1104. (e) Bica, K.; Gaertner, P. Org. Lett. 2006, 8, 733. (f)
Dongol, K. G.; Koh, H.; Sau, M.; Chai, C. L. L. AdV. Synth. Catal. 2007,
349, 1015. (g) Bedford, R. B.; Bruce, D. W.; Frost, R. M.; Hird, M. Chem.
Commun. 2005, 4161.
(29) Preliminary communication: Fu¨rstner, A.; Martin, R.; Majima, K. J. Am.
Chem. Soc. 2005, 127, 12236.
(30) (a) Jonas, K.; Schieferstein, L. Angew. Chem., Int. Ed. Engl. 1979, 18,
549. (b) Jonas, K.; Schieferstein, L.; Kru¨ger, C.; Tsay, Y.-H. Angew. Chem.,
Int. Ed. Engl. 1979, 18, 550. (c) Schieferstein, L.; Dissertation, Ruhr-
University, Bochum, 1978. (d) Klusmann, P., Dissertation, Ruhr-University,
Bochum, 1993.
(31) (a) Jonas, K. Angew. Chem., Int. Ed. Engl. 1985, 24, 295. (b) Jonas, K.;
Kru¨ger, C. Angew. Chem., Int. Ed. Engl. 1980, 19, 520. (c) Jonas, K. Pure
Appl. Chem. 1990, 62, 1169.
(23) For additional examples, see: (a) Taillefer, M.; Xia, N.; Ouali, A. Angew.
Chem., Int. Ed. 2007, 46, 934. (b) Fukuhara, K.; Urabe, H. Tetrahedron
Lett. 2005, 46, 603. (c) Bart, S. C.; Lobkovsky, E.; Chirik, P. J. J. Am.
Chem. Soc. 2004, 126, 13794. (d) Bart, S. C.; Bowman, A. C.; Lobkovsky,
E.; Chirik, P. J. J. Am. Chem. Soc. 2007, 129, 7212. (e) Bart, S. C.;
Hawrelak, E. J.; Lobkovsky, E.; Chirik, P. J. Organometallics 2005, 24,
5518.
9
J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008 1993

A R T I C L E S
Scheme 1 a
Fu¨rstner et al.
Figure 1. Molecular structure of 4 in the solid state (only one of the four
crystallographically independent molecules is depicted).33 Anisotropic
displacement parameter ellipsoids are shown at the 50% probability level,
hydrogen atoms have been omitted for clarity.
a (a) (i) Li, ethylene (1 bar), THF, -50 °C f 0 °C, (ii) N,N,N′,N′-
tetramethylethylene-diamine (TMEDA), -30 °C, 45% (23 g scale); (b) (i)
Li, THF, ethylene (5 bar), -80 °C f RT; (ii) TMEDA, Et2O, 0 °C, 43%
(12 g scale); (c) (i) Li, COD, THF, -50 °C; (ii) TMEDA, Et2O, RT, 75%;
(d) Li, COD, DME, -50 °C f RT, 97% (85 g scale) [50% after
recrystallization, 11 g scale)]; (e) Ph3CCl, pentane, -35 °C f RT, 70%.
DME adduct 4 become available. On treatment with 1,2-
dichloroethane or trityl chloride, complex 4 undergoes a
remarkably clean one-electron oxidation to give the formal
Fe(+1) complex 5 with a 17 electron count.32 The corresponding
half-sandwich complexes bearing Cp* (Cp* ) pentamethyl-
cyclopentadienyl) rather than Cp ligands are also accessible as
exemplified by compound 6.32 Detailed procedures for the large-
scale preparation of ferrates 1 and 4, serving as the preferred
catalysts in this study, are given in the Experimental Section of
this paper.
Figure 2. Molecular structure of 1 in the solid state. Anisotropic
displacement parameter ellipsoids are shown at the 50% probability level,
hydrogen atoms have been omitted for clarity.
in the related TMEDA complex 3 (2.530 Å).30 Further short
contacts are observed between the lithium atom and one of the
Cp-ring carbon atoms (2.466(13) Å) as well as one of the
two proximal olefinic sites of the COD ligand (2.33(4) Å);
the distance to the other double bond is substantially longer
(2.73(6) Å). These features indicate substantial back-bonding
of electron density from the central metal into the π* orbital of
the alkenes, which is also evident from the considerable
elongation of the CdC bond from a theoretical value of 1.34 Å
in free COD to an average of 1.436(10) Å in 4.33
A look at the spectroscopic data and X-ray structures of
representative ferrate complexes is highly informative. Thus,
the 13C NMR spectrum of 3 in toluene-d8 at low temperature
shows two signals each for the olefins (δ ) 52.5 (d), 41.6 (d)
ppm) and the methylene groups (δ ) 35.5 (t), 32.9 (t) ppm) of
the COD ligand, which coalesce if the spectrum is recorded at
+80 °C (δ ) 47.9 (d), 34.1 (t) ppm). This behavior is deemed
to reflect the increasing mobility of the Li cation on the anionic
[Fe] template, which is in accordance with the observed
conductivity of THF solutions of 3 at ambient temperature.30 It
is also interesting to note that formal replacement of the
TMEDA ligand on the lithium counterion by DME markedly
increases the intrinsic motion, because complex 4 shows only
one set of signals even at -78 °C.
The structure of 4 in the solid state (Figure 1) is composed
of four crystallographically independent molecules, which show
almost identical molecular conformations.33 Small but significant
differences of the lithium and iron coordination spheres are
revealed only upon closer inspection. All four independent
molecules are characterized by a remarkably close contact
between the iron center and the lithium atom which averages
2.493(4) Å and is significantly shorter than the Fe‚‚‚Li distance
Although the global structure of the corresponding ethylene
complex 134 in the solid state is similar, the interactions between
the lithium atom and the bound olefins become even more
prominent (Figure 2). Specifically, the lithium center resides
even closer to the ethylene moieties, with which it entertains
now almost equidistant contacts at 2.354(2) and 2.368(2) Å;
thereby, the Fe‚‚‚Li distance remains virtually unchanged
(2.5097(19) Å). Transpiring from these structural details is the
eminent role of the olefins for the stabilization of the electron
rich iron center in 1 and 4; yet, they can be readily displaced
by other ligands such as CO, which leads to the known half
sandwich complex 7 (M ) Li)35 in good yield.
(34) The structure of 1 was briefly discussed in the following review, but no
data have been deposited: Angermund, K.; Claus, K. H.; Goddard, R.;
Kru¨ger, C. Angew. Chem., Int. Ed. Engl. 1985, 24, 237.
(35) (a) Piper, T. S.; Wilkinson, G. J. Inorg. Nucl. Chem. 1956, 3, 104. (b)
King, R. B. Acc. Chem. Res. 1970, 3, 417. (c) Burlitch, J. M.; Ulmer, S.
W. J. Organomet. Chem. 1969, 19, P21. (d) Theys, R. D.; Vargas, R. M.;
Wang, Q.; Hossain, M. M. Organometallics 1998, 17, 1333 and literature
cited therein.
(32) Jonas, K.; Klusmann, P.; Goddard, R. Z. Naturforsch., B: Chem. Sci. 1995,
50, 394.
(33) For a depiction of the four independent molecules in the unit cell and a
more detailed discussion of the structure, see the Supporting Information.
9
1994 J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008

Cheap Metal for a “Noble” Task
A R T I C L E S
Scheme 2. Related Metalate Complexes
Scheme 3. Envisaged Role of Ferrate Complexes as Catalysts for
the Cycloisomerization of Enynes
Table 1. Screening of Various Ferrate Complexes and Related
Catalysts; E ) COOEt
The even more highly reduced tetraethyleneferrate(-2)
species 230 is reminiscent of Collman’s reagent [M2Fe(CO)4]
(8, M ) Li, Na, K);36,37 formally, the tightly bound and strongly
backbonding CO ligands of 8 are replaced by four ethylene units.
The olefins in 2, however, remain kinetically labile, thus
imparting a somewhat “bare” character onto the exceptionally
nucleophilic iron center of this complex. In structural terms,
however, complexes 2 and 8 show distinct differences: whereas
there is a substantial direct interaction of the Fe and the Li
entry
substrate
catalyst (5%)
conditions
t (h)
yield
1
2
3
4
5
6
7
8
11
11
11
11
11
11
11
11
11
11
13
13
13
13
13
1
2
3
4
5
0.1 M, 90 °C
0.1 M, 90 °C
0.1 M, 90 °C
0.1 M, 90 °C
0.1 M, 90 °C
0.1 M, 95 °C
0.1 M, 95 °C
0.1 M, 90 °C
0.1 M, 20 °C
0.1 M, 90 °C
0.1 M, 90 °C
0.1 M, 90 °C
0.08 M, reflux
0.02 M, reflux
0.01 M, reflux
6
12
12
12
12
12
12
12
30
12
18
18
4
83%
0%
82%
80%
<20%
0%
0%
0%
72%
0%
0%
centers in 2 according to the published crystal structure,30
8
consists of discrete [Fe(CO)4]2- fragments held together in the
solid state by a three-dimensional network of remote interactions
with the K+ counterions (Scheme 2).38
CpFe(CO)2I
[CpFe(CO)2]2
7
9
Iron Catalyzed Alder-Ene Reactions. The known propensity
of 1-4 to undergo ligand exchange spurred our efforts to use
such complexes as catalysts for cycloisomerization and cy-
cloaddition reactions.29,39 A recent investigation has shown that
alkynes bind significantly more tightly than alkenes to low-
valent iron centers.40 Therefore, we reasoned that the alkene
ligands in 1, 3, or 4 might be replaced by chelating substrates
containing at least one alkyne unit. If such an exchange process
allows one to bring, for example, 1,6-enynes into the coordina-
tion sphere of the electron rich lithium ferrate unit, an oxidative
cyclization might ensue (Scheme 3). As the metal center in 1
or 4 is isoelectronic with Rh(+1), the resulting metallacycle
might possibly evolve via a â-hydride elimination/reductive
elimination sequence, resulting in a net Alder-ene type cyclo-
isomerization of the substrate. Apart from a few remarkable
exeptions,21,41 Alder-ene reactions are commonly performed with
9
10
11
12
13
14
15
10
1
4
4
4
4
53%
60%
80%
83%
4
20
noble metal catalysts;39,42-44 if the hypothesis sketched in
Scheme 3 can be successfully reduced to practice, however,
the iron complexes in question might qualify as cheap,45
nontoxic, benign, and structurally well-defined alternatives that
are available in quantity.
Compounds 11 and 13 were chosen as model substrates to
optimize the reaction conditions. As shown in Table 1, the
Alder-ene product 12 was obtained in high yield after 6 h
reaction time in the presence of ferrate 1 (5 mol %) in toluene
at 80-90 °C. The analogous COD complexes 3 and 4 performed
(36) (a) Collman, J. P. Acc. Chem. Res. 1975, 8, 342. (b) Ellis, J. E. AdV.
Organomet. Chem. 1990, 31, 1.
(37) For other structurally defined iron ate complexes see the following for
leading references: (a) Brennessel, W. W.; Jilek, R. E.; Ellis, J. E. Angew.
Chem. Int. Ed. 2007, 46, 6132. (b) Do¨ring, M.; Uhlig, E. Z. Chem. 1986,
26, 449. (c) Rosa, P.; Me´zailles, N.; Ricard, L.; Mathey, F.; Le Floch, P.;
Jean, Y. Angew. Chem., Int. Ed. 2001, 40, 1251. (d) Ellis, J. E.; Chen, Y.
S. Organometallics 1989, 8, 1350.
(42) Leading references on Pd-based catalysts: (a) Review: Trost, B. M. Acc.
Chem. Res. 1990, 23, 34 and literature cited therein. (b) Trost, B. M.;
Lautens, M. J. Am. Chem. Soc. 1985, 107, 1781. (c) Hatano, M.; Terada,
M.; Mikami, K. Angew. Chem., Int. Ed. 2001, 40, 249.
(43) Leading references on Ru-based catalysts: (a) Trost, B. M.; Toste, F. D.
J. Am. Chem. Soc. 2002, 124, 5025. (b) Trost, B. M.; Toste, F. D. J. Am.
Chem. Soc. 1999, 121, 9728. (c) Trost, B. M.; Toste, F. D. J. Am. Chem.
Soc. 2000, 122, 714. (d) Trost, B. M.; Surivet, J.-P.; Toste, F. D. J. Am.
Chem. Soc. 2004, 126, 15592. (e) Trost, B. M.; Dong, L.; Schroeder, G.
M. J. Am. Chem. Soc. 2005, 127, 10259. (f) Chen, H.; Li, S. Organome-
tallics 2005, 24, 872.
(38) Teller, R. G.; Finke, R. G.; Collman, J. P.; Chin, H. B.; Bau, R. J. Am.
Chem. Soc. 1977, 99, 1104.
(39) Authoritative reviews: (a) Trost, B. M.; Toste, F. D.; Pinkerton, A. B.
Chem. ReV. 2001, 101, 2067. (b) Aubert, C.; Buisine, O.; Malacria, M.
Chem. ReV. 2002, 102, 813. (c) Trost, B. M.; Krische, M. J. Synlett 1998,
1. (d) Oppolzer, W. Angew. Chem., Int. Ed. Engl. 1989, 28, 38. (e) Fairlamb,
I. J. S. Angew. Chem., Int. Ed. 2004, 43, 1048. (f) Lautens, M.; Klute, W.;
Tam, W. Chem. ReV. 1996, 96, 49. (g) Wender, P. A.; Bi, F. C.; Gamber,
G. G.; Gosselin, F.; Hubbard, R. D.; Scanio, M. J. C.; Sun, R.; Williams,
T. J.; Zhang, L. Pure Appl. Chem. 2002, 74, 25.
(40) Yu, Y.; Smith, J. M.; Flaschenriem, C. J.; Holland, P. L. Inorg. Chem.
2006, 45, 5742.
(41) Low valent titanium: (a) Sturla, S. J.; Kablaoui, N. M.; Buchwald, S. L.
J. Am. Chem. Soc. 1999, 121, 1976. (b) Sato, F.; Urabe, H.; Okamoto, S.
Pure Appl. Chem. 1999, 71, 1511. (c) Takayama, Y.; Okamoto, S.; Sato,
F. J. Am. Chem. Soc. 1999, 121, 3559.
(44) Leading references on Rh-based catalysts: (a) Lei, A.; He, M.; Zhang, X.
J. Am. Chem. Soc. 2002, 124, 8198. (b) Oppolzer, W.; Fu¨rstner, A. HelV.
Chim. Acta 1993, 76, 2329.
(45) According to the current 2007-2008 Aldrich price list, 500 g of Cp2Fe
costs 132 Euros; if purchased on a larger scale, it may actually be
significantly cheaper because ferrocene is a bulk chemical used as fuel
additive in certain countries. For a very cost effective and scaleable
synthesis, see: Eisenbach, W.; Lehmkuhl, H. Chem. Ing. Tech. 1982, 54,
690. For comparison, 1 g of Pd(OAc)2 as the cheapest palladium salt
presently costs 108 Euros, whereas 1 g of (Ph3P)3RhCl amounts to 112
Euros.
9
J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008 1995

A R T I C L E S
Fu¨rstner et al.
similarly well, although a somewhat longer reaction time was
necessary (12 h). This rate difference is deemed to reflect the
more facile and irreversible substitution of the ethylene ligands
in 1 by the enyne, whereas the chelating COD in 3 or 4 is
arguably more difficult to replace, remains in solution and might
therefore compete with substrate binding. Although this kinetic
retardation may seem undesirable at first sight, the presence of
COD in the reaction medium actually turns out to be advanta-
geous with somewhat more demanding substrates. For example,
compound 13 would not cyclize with ferrate 1 (entry 11),
whereas the COD complex 4 furnished the desired product 14
in up to 83% yield. This striking difference is tentatively
ascribed to a stabilizing effect of COD, which may serve as
ancillary ligand to protect the catalyst in the resting state. As
evident from entries 12-15, Alder-ene reactions of acyclic
substrates such as 13 are best performed in more dilute solution
(0.01-0.05 M).
In line with the notion that the olefins in precatalysts 1 or 4
are initially replaced by the enyne substrate when entering the
catalytic cycle, formal exchange of these ligands by strongly
bound CO (complex 7, M ) Na, entry 8) results in complete
loss of catalytic activity. The neutral iron carbonyl precursors
[CpFe(CO)2I] and [CpFe(CO)2]2 were equally unsuitable (entries
6-7). The fact that the more highly reduced tetraethylene ferrate
2 failed to afford the desired product 12 is tentatively ascribed
to the lability of this complex at higher temperatures (entry 2).
It is interesting to note, however, that the corresponding ethylene
cobaltate complex [(C2H4)4Co]K (9)31,46 is catalytically com-
petent even at ambient temperature, although the reaction needed
30 h to go to completion (entry 9).
Another noteworthy observation concerns the analogous
ruthenate complex 10,47 which was found inactive (entry 10).
Even though this species contains a 1,3-diene rather than a 1,5-
diene ligand, the comparison with the closely related iron
complex 4 showcases the much higher catalytic competence of
low valent iron in the present context. The failure of the
nucleophilic Ru(0) species 10 is somewhat surprising if one
considers the exceptional utility of electrophilic complexes of
ruthenium in higher oxidation states as catalysts for Alder-ene
reactions and related processes.39,43,48
The two optimized reaction conditions (A: 1 cat., toluene
0.1 M, 90 °C; B: 4 cat., toluene, 0.01 M, reflux) were then
applied to a host of different substrates to investigate the scope
and limitations of this new iron catalyzed Alder-ene protocol.
Several observations deserve comment: Although the formation
of bicyclic products by Alder-ene processes is rare, the iron-
based method is particularly well suited for such annulation
reactions (Table 2). The pre-existing ring of the starting material,
however, determines the stereochemistry at the ring junction
(see below) and exerts a distinct influence on the reaction rate,
with larger cycloalkenes usually rendering the transformation
more facile. This is evident from the fact that the [3.3.0]bicyclo-
octene derivative (entry 1) was the most difficult to form among
all products compiled in Table 2. It is believed that the strain
in the metalla-tricyclic intermediate peaks for this particular
compound (see below).
Except for the decaline product derived from a 1,7-enyne
(entry 4), which is formed as a cis/trans-mixture, all compounds
shown in Table 2 were obtained as single isomers. Thereby,
the exocyclic alkene is invariably (E)-configured as expected
for an Alder-ene pathway. The newly generated endocyclic
olefin must be (Z)-configured when part of a 5-8 membered
ring (entries 1-18), whereas it is the corresponding (E)-isomer
that is exclusively formed wherever the ring size allows to (g10,
entries 19-30).49
Although one might be concerned about the potential basicity
of ferrate complexes in general, enynes containing terminal
acetylene units posed no problems (entries 6, 19, 24). Likewise,
different substituents on the alkyne are well accommodated,
including electron-withdrawing substituents, cyclopropyl- and
silyl groups. The latter result is remarkable because silylated
enynes are unsuitable for Alder-ene reactions catalyzed by low-
valent titanium reagents.41a If the alkyne carries an aryl
substituent, the electronic characteristics of this ring are largely
inconsequential. Particularly significant is the compatibility of
the iron catalyst with various tethers and functional groups,
including esters, ketones, acetals, ethers, silyl ethers, sulfon-
amides, pyrroles, cyclopropanes, and remote alkenes; even a
tertiary amine does not interfere, indicating a moderate Lewis
acidity of the chosen catalyst system, if any (entry 13). The
fact that aryl halides also remain untouched shows that the
oxidative cyclization of the enyne must be favored over a
conceivable oxidative insertion of the low-valent iron center
into such C-X bonds (entries 14, 27).
Table 3 compiles representative examples of iron-catalyzed
Alder-ene reactions of acyclic enynes according to Scheme 4.
As already mentioned above, the substitution pattern of such
substrates has a pronounced effect on their reactivity. Specif-
ically, enynes with R2 ) H failed to react with the bis-ethylene
complex 1 under the standard conditions, whereas they were
cleanly converted to the desired Alder-ene products with the
aid of the analogous COD complex 4 (conditions B) (cf. entries
1/2, 3/4, 5/6). This superior performance of 4, which also
transpires upon comparison of entries 8/9 and 20/21, is thought
to reflect the stabilizing influence of the ancillary COD ligand
remaining in solution, which was also noticed during our
investigations on the [5+2] cycloaddition processes (vide infra).
In striking contrast, conformationally more biased enynes with
R2 * H react with both catalysts, affording the corresponding
trans-disubstituted products as the major isomers in all cases
investigated.50 As in the annulation mode discussed above, the
reaction applies to 1,6- as well as 1,7-enynes, tolerates various
kinds of tethers, and is compatible with basic amine functions
as well as various potentially reducible functional groups.
The major limitation encountered during our investigation
concerns enynes bearing a trisubstituted alkene moiety that failed
to react under the chosen conditions. Representative examples
of such unreactive substrates are compiled in Chart 1.
Stereochemical Aspects and Mechanistic Investigation. A
striking aspect of the Alder-ene annulation process concerns
(49) The endocyclic olefin was invariably (E)-configured even though the
substrates used were E/Z-mixtures. Moreover, GC investigations show that
the (E)-isomer does not result from equilibration but is the primary product
of the reaction.
(50) The stereochemical assignment is based on NOESY data [Table 2: entry
1 (cis), entry 2 (cis), entry 4 (trans), entry 6 (trans), entry 27 (trans); Table
3: entry 16 (trans), entry 28 (trans)]; all other products were assigned by
analogy.
(46) See also: Jonas, K.; Mynott, R.; Kru¨ger, C.; Sekutowski, J. C.; Tsay,
Y.-H. Angew. Chem., Int. Ed. Engl. 1976, 15, 767.
(47) Fagan, P. J.; Mahoney, W. S.; Calabrese, J. C.; Williams, I. D. Organo-
metallics 1990, 9, 1843.
(48) Review: Trost, B. M.; Frederiksen, M. U.; Rudd, M. T. Angew. Chem.,
Int. Ed. 2005, 44, 6630.
9
1996 J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008

Cheap Metal for a “Noble” Task
A R T I C L E S
Table 2. Iron Catalyzed Alder-ene Annulation Reactions;a E ) COOEt
a All reactions were performed with complex 1 (5 mol %) in toluene at 90 °C, unless stated otherwise. b Using 10 mol % of catalyst. c dr ≈ 2.5:1. d Using
20 mol % of catalyst.
its stereodivergent course with regard to the ring junction.
Whereas enynes containing a cyclopentene- or cyclohexene ring
exclusively form the corresponding cis-annelated ring systems
that a feeble yet non-negligible NOE is recorded between the
protons at the ring junction even though they are trans disposed.
Although the overall NMR data set leaves no doubt with regard
to the stereochemical assignment, the trans-ring fusion was
independently confirmed for sulfonamide 15 (Table 2, entry 11)
by X-ray structure analysis to avoid any ambiguity in this regard
(Figure 3).
3
(Table 2, entries 1-3) as evident from the pertinent JH3,H4
coupling constants and the strong NOE effects between these
protons (Scheme 5), all substrates incorporating larger cyclo-
alkenes uniformly lead to the corresponding trans-adducts (Table
2, entries 5-30).51
This assignment is based on the observed large coupling
constant (3JH3,H4 ≈ 11 Hz) and the analysis of the entire NOE
pattern indicated in Scheme 5. It is important to note, however,
(51) Some precedence for a related stereochemical switch from cis to trans on
going to larger ring sizes is found in organopalladium chemistry, cf: Trost,
B. M.; Grese, T. A. J. Am. Chem. Soc. 1991, 113, 7363.
9
J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008 1997

A R T I C L E S
Fu¨rstner et al.
Table 3. Iron Catalyzed Alder-ene Type Reactions of Acyclic
Enynes;a E ) COOEt
Chart 1
Scheme 5. Stereochemical Assignment of Representative cis-
and trans-fused Alder-ene Products
trajectory as exemplified for the substrate class 20 incorporating
a cyclooctenyl entity (Scheme 6, bottom). To avoid a clash with
the hydrogen in transannular proximity, the alkyne in the tether
approaches the olefin in a pseudo-equatorial fashion, which
readily explains the experimentally observed trans-annulation
mode in this and related cases.
This interpretation tacitly assumes metallacycles as the key
reactive intermediates, which, however, is by no means the only
conceivable pathway. Metal-catalyzed Alder-ene reactions in
general may occur by one of three different scenarios which
a Conditions: A: complex 1 (5 mol %), toluene (0.1 M), 90 °C; B:
complex 4 (5 mol %), toluene (0.02 M), reflux. b Only the major isomer is
depicted. c Using 10 mol % of catalyst. d Using 20 mol % of catalyst.
e Using 15 mol % of catalyst.
Scheme 4. Iron-catalyzed Reactions of Acyclic Enynes
This stereodivergent course of the reaction likely reflects
distinct conformational preferences of the different ring sizes.
We assume that substrates of type 16 adopt a transition state
placing the tether in a pseudoequatorial position on the fairly
rigid cyclohexenyl scaffold; an ensuing axial attack then
generates metallacycle 18 and hence nicely explains the
formation of the cis-annelated bicycle 19 as the only product
(Scheme 6). In contrast, transannular interactions dominate the
behavior of medium rings, which strongly disfavor such a
Figure 3. Molecular structure of 15 (Table 2, entry 11) in the solid state.
Anisotropic displacement parameter ellipsoids are shown at the 50%
probability level, hydrogen atoms have been omitted for clarity, except those
at the ring junction.
9
1998 J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008

Cheap Metal for a “Noble” Task
A R T I C L E S
Scheme 6. Tentative Explanation for the Stereodivergent Course
of the Alder-ene Annulation Process as a Function of the Size of
the Pre-existing Ring
to generate the required metal hydride initiator, the available
data do not allow for a clear-cut choice between the two
remaining options. We reasoned, however, that the stereodi-
vergent course of the iron-catalyzed Alder-ene annulation
process may provide a unique opportunity to gather indirect
evidence via labeling experiments.
If the reaction of the stereospecifically labeled cyclooctenyl
derivative 24 commences with allylic CH activation, the iron
catalyst will insert into the C-H as well as into the C-D bond
with similar ease (Scheme 8, right).54-56 Activation of the C-D-
bond affords the allyliron species Ia which transfers the label
to the exocyclic alkene of trans-annulated bicycle as the only
observed product (24 f 25a).57 However, the isomeric allyliron
complex Ic, generated by activation of the allylic C-H rather
than C-D bond, can also lead to the trans-annulated product if
an allyl-interconversion to Id intervenes. Such a process may
be triggered by π-σ-π hapticity changes or result from
intermolecular metal exchange, which are well precedented
phenomena in allyliron chemistry.58-60 Because a course via
Ic,d transmits hydrogen but leaves the label intact at its original
site, it delivers isotopomer 25b. One may hence anticipate that
a mechanism based on CH-activationsoverallslikely results
in some degree of scrambling of the deuterium in the product
(25a/25b).
In contrast, a metallacyclic pathway can afford the experi-
mentally observed trans-annulated product only via an inter-
mediate of type II as a consequence of the geometrical
constraints inherent to a tricyclic scaffold (Scheme 8, left). In
this case, â-elimination exclusively activates the deuterium label
in cis-disposition relative to the C-Fe bond. Hence, a metal-
lacyclic scenario must engender a complete transmission of the
label to the exocyclic olefin (II f III f 25a only).
Scheme 7. Conceivable Mechanistic Scenarios of
Metal-Catalyzed Alder-ene Reactions of Enynes
The same arguments apply to the formation of the cis-
annulated product 29 derived from the labeled cyclohexenyl
(54) Even if one assumes that this step is rate determining and hence a kinetic
isotope effect comes into play, the probabilities for activation of the C-H
and the C-D bonds are of the same order of magnitude.
(55) For the facile formation of allyl- and pentadienyliron complexes by allylic
CH activation, see the following for a leading reference: Angermund, K.;
Geier, S.; Jolly, P. W.; Kessler, M.; Kru¨ger, C.; Lutz, F. Organometallics
1998, 17, 2399.
(56) For evidence for fast allylic CH activation by low valent iron in the gas
phase as the first step of a formal Alder-ene type process, see: Huang, Y.;
Freiser, B. S. J. Am. Chem. Soc. 1990, 112, 1682.
(57) Note that reductive elimination from Ib would afford the cis-adduct which
is not observed; therefore, putative Ib must be outside the productive
pathway.
(58) (a) The π f σ hapticity change is particularly fast upon coordination of
an extra ligand to a π-allyliron species; the alkyne could play such a role
in the present case, see: (a) Gabor, B.; Holle, S.; Jolly, P. W.; Mynott, R.
J. Organomet. Chem. 1994, 466, 201. (b) Cardaci, G.; Foffani, A. J. Chem.
Soc., Dalton Trans. 1974, 1808. (c) Hapticity change accompanied by an
exo/endo rearrangement: Faller, J. W.; Adams, M. A. J. Organomet. Chem.
1979, 170, 71.
(59) Reviews: (a) π-allyliron complexes: Moss, J. R.; Smith, G. S.; Kaschula,
C. H. In ComprehensiVe Organometallic Chemistry III; Mingos, D. M. P.,
Crabtree, R. H., Eds.; Elsevier: Amsterdam, 2007; Vol. 6, pp 127. (b)
σ-Allyliron complexes: Knorr, M. In ComprehensiVe Organometallic
Chemistry III; Mingos, D. M. P., Crabtree, R. H., Eds.; Elsevier:
Amsterdam, 2007; Vol. 6; pp 77.
(60) Less common than π-allyliron complexes, σ-allyliron species are well
known and have been isolated and fully characterized in certain cases.
Additionally, there is strong evidence for various reactions to proceed via
σ-allyliron complexes as the key intermediates. For leading references see
ref. 17 and the following, together with literature cited therein: (a) Akita,
M.; Shirasawa, N. Chem. Commun. 1998, 973. (b) Akita, M. J. Organomet.
Chem. 2004, 689, 4540. (c) Eberhardt, U.; Mattern, G. Chem. Ber. 1988,
121, 1531. (d) Stereochemical arguments for σ-allyliron: Xu, Y.; Zhou,
B. J. Org. Chem. 1987, 52, 974. (e) Benn, R.; Brenneke, H.; Frings, A.;
Lehmkuhl, H.; Mehler, G.; Rufinska, A.; Wildt, T. J. Am. Chem. Soc. 1988,
110, 5661. (f) Rosenblum, M. Acc. Chem. Res. 1974, 7, 122. (g) Dey, K.;
Koner, D.; Bhattacharyya, P. K.; Gangopadhyay, A.; Bhasin, K. K.; Verma,
R. D. Polyhedron 1986, 5, 1201. (h) Bassett, J.-M.; Green, M.; Howard, J.
A. K.; Stone, F. G. A. J. Chem. Soc., Dalton Trans. 1980, 1779.
can be classified according to the initial step as follows (Scheme
7):39,52 (1) addition of a metal hydride species formed in situ or
ex situ to the alkyne generates a vinyl metal reagent which adds
to the double bond in a “Heck type” manner and thereby leads
to the final product; (2) activation of the allylic CH-bond by
the catalyst generates an allylmetal species which undergoes a
“metallo-ene” type reaction with the alkyne; (3) oxidative
cyclization of the enyne generates a metallacyclopentene, which
affords the product via a subsequent â-hydride elimination/
reductive elimination process.
A priori, it is very difficult to rigorously distinguish between
these scenarios,53 and direct spectroscopic evidence is usually
lacking due to the very limited lifetimes of the putative
intermediates.39 Whereas the metal hydride mechanism can
likely be ruled out in our case, since no proton source is present
(52) Review on mechanisms of cycloisomerization reactions of enynes: Lloyd-
Jones, G. C. Org. Biomol. Chem. 2003, 1, 215.
(53) (a) In this context, it is interesting to note that most Ru-catalyzed Alder-
ene reaction were analyzed on the basis of metallacyclic intermediates;
however, a mechanistic switch to a CH activation pathway was proposed
to explain certain outcomes, cf. ref 43. (b) A CH activation process was
also invoked in closely related cobalt-catalyzed Alder-ene processes, cf.:
Ajamian, A.; Gleason, J. L. Org. Lett. 2003, 5, 2409.
9
J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008 1999

A R T I C L E S
Fu¨rstner et al.
Scheme 8. Deuterium Labeling Experiments as Possible Probes for the Mechanism of the Iron Catalyzed Alder-ene Reaction of Substrate
24
Scheme 9. Stereochemical Implication of a Metallacyclic Pathway
for the Alder-ene Reaction of the Cyclohexenyl Derivative 26
Scheme 10
product 32 was obtained.62 This outcome is general, although
the yields of the cycloadducts were only moderate (54-67%).
Control experiments confirmed the catalytic role of the iron
species, which accelerates the [4+2] cycloaddition over the
thermal background reaction (Table 4). Because metal-catalyzed
Diels-Alder reactions of unactivated alkynes as dienophiles are
thought to proceed via metallacyclic intermediates,39,63 this
behavior is consistent with the mechanistic conclusions reached
above.
derivative 26. An allylic CH-activation pathway will again likely
engender some degree of scrambling. It is of special note,
however, that a metallacyclic intermediatesin striking contrast
to the cyclooctenyl case discussed abovesnecessarily invokes
a quantitative preserVation of the deuterium atom at the original
site as evident from Scheme 9.
The experimental results are highly informative (Scheme 10).
Specifically, reaction of 2461 with catalytic amounts of complex
1 in toluene at 90 °C gave 25a as the only product in 90%
isolated yield, in which the deuterium label has quantitatively
migrated to the exocyclic alkene. Under the exact same
conditions, the smaller homologue 26, in contrast, exclusively
affords compound 29 (82%) in which the deuterium atom is
completely retained at the original site. Within the limits of
detection, no sign of scrambling of the label was observed in
either case. Even though these opposite labeling patterns cannot
be taken as an ultimate proof for a metallacyclic pathway, they
do provide strong evidence for such a scenario.
Iron Catalyzed [5+2] Cycloaddition Reactions. The success
of the iron-catalyzed Alder-ene and Diels-Alder reactions
encouraged us to extend our program to other skeletal reorga-
nization processes. Given the fact that [5+2] cycloadditions are
not only exceedingly useful for the construction of variously
substituted seven-membered rings but also dominated by noble
metal catalysis,48,64-66 such transformations constitute another
(61) The deuterated cyloalkenes were prepared as described in: (a) Franze´n,
J.; Ba¨ckvall, J.-E. J. Am. Chem. Soc. 2003, 125, 6056. (b) Ba¨ckvall, J.-E.;
Nystro¨m, J.-E.; Nordberg, R. E. J. Am. Chem. Soc. 1985, 107, 3676; for
details, see the Supporting Information.
(62) For precedence on Diels-Alder reactions catalyzed by structurally unknown
low-valent iron species generated in situ, see ref 19a.
(63) For alternative mechanisms allowing [4+2] cycloadditions of diene-ynes
to proceed, see: Fu¨rstner, A.; Stimson, C. C. Angew. Chem., Int. Ed. 2007,
46, 8845.
Iron Catalyzed [4+2] Cycloadditions. Diene-yne derivatives
such as 30 were previously shown to undergo unusual Alder-
ene reactions to give ene-allenes of type 31 in the presence of
low valent titanium.41a,21 When reacted with catalytic amounts
of 4 under the usual conditions, however, only the Diels-Alder
9
2000 J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008

Cheap Metal for a “Noble” Task
A R T I C L E S
Table 4. Intramolecular [4+2] Cycloadditions of Unactivated
Diene-ynes Catalyzed by the Iron-ate Complex 4;a E ) COOEt
Table 5. [5+2] Cycloaddition Reactions Catalyzed by Ferrate(0)
Complexes;a E ) COOEt
entry
X
R
catalyst
t (h)
conversion
yieldb
1
2
3
4
5
6
7
8
NTs
H
H
H
H
Me
Me
-
1.5
1.5
7
23%
>90%
25%
85%
10%
100%
25%
100%
4 (10%)
-
54%
67%
57%
66%
CE2
4 (10%)
-
7
16
16
6
4 (20%)
-
SiMe3
SiMe3
4 (10%)
6
a All reactions were performed in toluene at 80-90 °C. b After work up
in air, the products contain variable amounts of the fully aromatized
compounds.
Scheme 11. Iron-Catalyzed [5+2] Cycloadditions
attractive testing ground for the cheap ferrate complexes
discussed herein (Scheme 11).
a Conditions: A: complex 1 (10 mol %), toluene (0.1 M), 90 °C; B:
complex 4 (10 mol %), toluene (0.02 M), reflux. b Only the major isomer
is depicted. c No product was obtained under conditions A. d Using 20 mol
% of catalyst. e Using 5 mol % of catalyst. f Using 15 mol % of catalyst.
In line with our expectations, a set of representative vinyl-
cyclopropane derivatives were converted into the corresponding
cycloheptadiene derivatives on exposure to the Fe(0) complexes
1 or 4 under the established conditions (Table 5). Like in the
Alder-ene case, the COD-containing species 4 turned out to be
more generally applicable, allowing conformationally less biased
substrates with R2 ) H to be transformed to the corresponding
cycloadducts, whereas the ethylene complex 1 failed in such
cases. The products are obtained with good to excellent
diastereoselectivity, invariably favoring the 1,2-trans-disubsti-
tuted isomer.67 Terminal and differently end-capped alkynes
participate with similar ease (R1 ) H, Me, SiMe3, COOEt,
aryl), independent of the electronic variations that such sub-
stituents impose. Entry 8 shows that a [5.4.0] bicyclic product
can also be formed by this methodology. As in the Alder-ene
regimen, the only serious limitation concerns substrates with
trisubstituted olefins which remain unchanged under the chosen
conditions.
(64) Rhodium catalysis: (a) Wender, P. A.; Takahashi, H.; Witulski, B. J. Am.
Chem. Soc. 1995, 117, 4720. (b) Yu, Z.-X.; Wender, P. A.; Houk, K. N. J.
Am. Chem. Soc. 2004, 126, 9154. (c) Wender, P. A.; Sperandio, D. J. Org.
Chem. 1998, 63, 4164. (d) Wender, P. A.; Dyckman, A. J.; Husfeld, C. O.;
Kadereit, D.; Love, J. A.; Rieck, H. J. Am. Chem. Soc. 1999, 121, 10442.
(e) Wender, P. A.; Dyckman, A. J. Org. Lett. 1999, 1, 2089. (f) Wender,
P. A.; Williams, T. J. Angew. Chem., Int. Ed. 2002, 41, 4550. (g) Wender,
P. A.; Haustedt, L. O.; Lim, J.; Love, J. A.; Williams, T. J.; Yoon, J.-Y. J.
Am. Chem. Soc. 2006, 128, 6302. (h) Wender, P. A.; Glorius, F.; Husfeld,
C. O.; Langkopf, E.; Love, J. A. J. Am. Chem. Soc. 1999, 121, 5348. (i)
Wender, P. A.; Husfeld, C. O.; Langkopf, E.; Love, J. A. J. Am. Chem.
Soc. 1998, 120, 1940. (j) Ashfeld, B. L.; Martin, S. F. Tetrahedron 2006,
62, 10497. (k) Wender, P. A.; Rieck, H.; Fuji, M. J. Am. Chem. Soc. 1998,
120, 10976. (l) Wender, P. A.; Bi, F. C.; Brodney, M. A.; Gosselin, F.
Org. Lett. 2001, 3, 2105. (m) Saito, A.; Ono, T.; Hanzawa, Y. J. Org.
Chem. 2006, 71, 6437. (n) Lee, S. I.; Park, S. Y.; Park, J. H.; Jung, I. G.;
Choi, S. Y.; Chung, Y. K.; Lee, B. Y. J. Org. Chem. 2006, 71, 91.
(65) For a C-H activation/[5+2] tandem, see: A¨ıssa, C.; Fu¨rstner, A. J. Am.
Chem. Soc. 2007, 129, 14836.
Iron Catalyzed [2+2+2] Cycloadditions and Additional
Mechanistic Insights. Finally, the applicability of ferrate
complexes to catalytic [2+2+2] cycloadditions was briefly
surveyed. Such transformations are effected by a host of
transition metal catalysts, among which CpCoL2 (L ) CO,
C2H4) and various Rh(+1) complexes are arguably most widely
used.39,68,69 Because the ferrate complexes 1 and 4 are isoelec-
tronic with such templates, one might expect them to be
competent catalysts for such purposes too.
(67) The stereochemical assignment is based on NOESY data for compounds
shown in entries 9 and 11; all other products were assigned by analogy.
(68) (a) Vollhardt, K. P. C. Angew. Chem., Int. Ed. Engl. 1984, 23, 539. (b)
Bo¨nnemann, H.; Brijoux, W. AdV. Heterocycl. Chem. 1990, 48, 177. (c)
Varela, J. A.; Saa´, C. Chem. ReV. 2003, 103, 3787. (d) Yamamoto, Y.
Curr. Org. Chem. 2005, 9, 503. (e) Chopade, P. R.; Louie, J. AdV. Synth.
Catal. 2006, 348, 2307.
(66) Ruthenium catalysis: (a) Trost, B. M.; Toste, F. D.; Shen, H. J. Am. Chem.
Soc. 2000, 122, 2379. (b) Trost, B. M.; Shen, H. C. Org. Lett. 2000, 2,
2523. (c) Trost, B. M.; Shen, H. C.; Schulz, T.; Koradin, C.; Schirok, H.
Org. Lett. 2003, 5, 4149. (d) Trost, B. M.; Shen, H. C. Angew. Chem., Int.
Ed. 2001, 40, 2313. (e) Trost, B. M.; Toste, F. D. Angew. Chem., Int. Ed.
2001, 40, 1114 and literature cited therein.
9
J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008 2001

A R T I C L E S
Scheme 12 a
Fu¨rstner et al.
a Reagents and conditions: (a) complex 4 (10 mol %), toluene (0.08
M), reflux, 2 h, 89%; (b) complex 6 (20 mol %), toluene (0.08 M), reflux,
21 h, 80%; (c) complex 4 (5 mol %), toluene (0.1 M), 90 °C, 24 h, 70%;
E ) COOEt.
Figure 4. Molecular structure of 38 in the solid state. Anisotropic
displacement parameter ellipsoids are shown at the 50% probability level,
hydrogen atoms have been omitted for clarity.
In fact, triyne 33 was smoothly converted into product 34 on
exposure to catalytic amounts of 4 in refluxing toluene;
analogously, diyne 35 underwent a clean dimerization to give
36 in respectable yield (Scheme 12). Interestingly, the Fe(+1)
complex 6 was also active, although a higher loading and a
longer reaction time were needed. In each case, addition of the
substrate to the solution of the iron species was accompanied
by an instantaneous and very characteristic color change from
red to bright green. Although similar observations were made
with the other types of substrates discussed above, it was
particularly pronounced in this series. Speculating that this effect
might either reflect the exchange of the COD ligand for the
substrate or even the formation of the putative metallacyclic
intermediate on the reaction coordinate, attempts were made to
isolate and characterize the new species formed in situ.
Scheme 13. Possible Pathway for the Formation of the
17-electron Fe-Cyclobutadiene Complex 38
To this end, the different iron complexes were reacted with
stoichiometric amounts of 1,2-diphenylacetylene (tolane) as the
model substrate. Although all attempts to characterize the
corresponding complexes derived from this alkyne and 1 or 4
have so far been unsuccessful, the Fe(+1) complex 6 allowed
us to obtain a stable product in analytically pure form in 77%
yield. Spectroscopic data and elemental analysis showed that
two equivalents of tolane must have reacted with 6,30d but the
identity of this new compound could only be revealed by X-ray
structure analysis. As can be seen from Figure 4, it turned out
to be the Fe(+1) cyclobutadiene complex 38, a remarkable
compound given the odd electron count of the central metal.70
This 17-electron iron species is somewhat reminiscent of the
closed shell (CO)3Fe(0)(cyclobutadiene) complexes formed by
oxidative insertion of Collman’s reagent into cis-3,4-carbonyl-
dioxycyclobutenes, which, however, benefit from additional
stabilization by the strongly backbonding carbonyl ligands.71
Moreover, the formation of 38 from 6 parallels the behavior of
[Cp*Co(C2H4)2], which is a particularly effective catalyst for
[2+2+2] cycloadditions of various alkynes, yet transforms
tolane into the cyclobutadiene complex [Cp*Co(η4-C4Ph4)].72,73
The formation of 38 is best explained by assuming an
oxidative cyclization with formation of metallacycle 37 as the
primary product, which, upon reductive elimination, generates
the butadiene entity that remains bound to the regenerated
Fe(+1) template (Scheme 13). As the latter step is likely driven
by the release of strain caused by the clash of the bulky Cp*
ligand with the adjacent phenyl rings in 37, we suspected that
smaller substituents on the alkyne might allow one to intercept
other intermediates. Therefore, 6 was reacted with a slight excess
of 2-butyne which also resulted in the formation of an iron
complex that could be isolated in analytically pure form, even
though it is much more labile. Although no crystals suitable
for X-ray structure analysis could be obtained, all available data
suggest that the new iron complex 41 is the product of the
[2+2+2] cycloaddition between two molecules of butyne and
one ethene ligand derived from the starting complex itself, which
(69) For previous examples of iron-catalyzed cyclotrimerizations, see the
following for leading references and literature cited therein: (a) Breschi,
C.; Piparo, L.; Pertici, P.; Caporusso, A. M.; Vitulli, G. J. Organomet.
Chem. 2000, 607, 57. (b) Funhoff, A.; Scha¨ufele, H.; Zenneck, U. J.
Organomet. Chem. 1988, 345, 331. (c) Usieli, V.; Victor, R.; Sarel, S.
Tetrahedron Lett. 1976, 17, 2705. (d) Hu¨bel, W.; Hoogzand, C. Chem.
Ber. 1960, 93, 103. (e) Carbonaro, A.; Greco, A: Dall’Asta, G. J. Org.
Chem. 1968, 33, 3948. (f) Weber, S. R.; Brintzinger, H. H. J. Organomet.
Chem. 1977, 127, 45. (g) Simons, L. H.; Lagowski, J. J. J. Organomet.
Chem. 1983, 249, 195. (h) Saino, N.; Kogure, D.; Okamoto, S. 0rg. Lett.
2005, 7, 3065.
(70) The geometry of this Cp*Fe(C4Ph4) radical differs only slightly from that
of the anion in [Cp*Fe(C4Ph4)]- [Li(thf)4]+, whose structure was briefly
reported in the following publication: Kru¨ger, C.; Sekutowski, J. C.; Tsay,
Y.-H. Z. Kristallogr. 1979, 149, 109. Despite the higher uncertainty
associated with the earlier measurement, the tendency of the ligands to be
closer to the Fe atom in the anionic complex is evident.
(71) (a) Grubbs, R. H. J. Am. Chem. Soc. 1970, 92, 6693. (b) Grubbs, R. H.;
Grey, R. A. J. Chem. Soc., Chem. Commun. 1973, 76.
(72) Beevor, R. G.; Frith, S. A.; Spencer, J. L. J. Organomet. Chem. 1981,
221, C25-C27.
(73) Crystallographic investigation on the mechanism of Co-catalyzed, cyclo-
trimerizations: Diercks, R.; Eaton, B. E.; Gu¨rtzgen, S.; Jalisatgi, S.;
Matzger, A. J.; Radde, R. H.; Vollhardt, K. P. C. J. Am. Chem. Soc. 1998,
120, 8247.
9
2002 J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008

Cheap Metal for a “Noble” Task
A R T I C L E S
Scheme 14. Participation of an Ethylene Ligand of the 17-electron
Fe(+1) Complex 6 in a Mixed [2+2+2] Cycloaddition
the solid state are reported, which reveal the capacity of the
metal center to share electron density with the ligand environ-
ment and hence explains the capacity of such catalysts to induce
oxidative cyclization reactions of polyunsaturated compounds.
As a result, these complexes effect a variety of cycloaddition
and cycloisomerization reactions of the Alder-ene, [4+2], [5+2],
and [2+2+2] type. These transformations are compatible with
a variety of substitution patterns, potentially reducible functional
groups, basic sites, and even fairly CH-acidic terminal alkyne
units. Labeling experiments together with the isolation of two
novel Fe(+1) complexes, which incorporate different primary
cycloadducts, provide compelling evidence for the proposed
mechanism. Although additional investigations are warranted
to expand the scope and upgrade the catalytic performance of
the catalysts even further, we believe that this study constitutes
an important step toward a more prolific use of inexpensive,
readily available, nontoxic, and benign iron complexes as
powerful alternatives to the commonly used noble metal
catalysts in the realm of cycloaddition and cycloisomerization
chemistry.
Scheme 15 a
Experimental Section
The complete experimental section is found in the Supporting
Information. Outlined below is the preparation of the ferrate complexes
used as catalysts herein, as well as the synthesis of the novel 17-electron
iron cyclobutadiene complex 38.
Preparation of [CpFe(C2H4)2] [Li(tmeda)] (1). A flame-dried three-
necked round-bottom flask equipped with a magnetic stirbar, a gas inlet,
a glass stopper, and a stopcock connected to the vacuum line is charged
under Ar with lithium sand (2.40 g, 346 mmol) and THF (200 mL).
The suspension is cooled to -50 °C and vacuum is applied until gentle
boiling of the solvent is observed. At that stage, the flask is back-
filled with ethene (from a gas burette to monitor the uptake) and the
evacuation/back-filling cycle is repeated three times. Ferrocene (32.1
g, 173 mmol) is then added in portions under a gentle stream of ethene
and the resulting orange-brown suspension is vigorously stirred for
20 h at -50 °C. During the first 2-3 h, a fast uptake of ethylene can
be observed, which slowly ceases. The resulting mixture is allowed to
warm to 0 °C before it is filtered under Ar through a cooled funnel
(cooling jacket) at 0 °C. The filtrate is evaporated under reduced
pressure to ca. 1/2 of the original volume before freshly distilled
N,N,N′,N′-tetramethyl-ethylenediamine (TMEDA, 100 mL) is added.
Storing of the resulting solution at -30 °C overnight causes the
precipitation of orange-red crystals, which are filtered off, washed
with cold Et2O (0 °C, 200 mL in two portions), and dried at ambient
temperature under vacuum (10-3 Torr). The resulting air-sensitive
orange-red crystals of complex 1 (23.5 g, 45%) can be handled at
ambient temperature without noticeable decomposition and can be
stored at -20 °C for extended periods of time (>1 year). Although
the signals in the 1H NMR spectrum are broad, the 13C NMR recorded
at 193 K shows characteristic sharp lines. 13C NMR (100 MHz, THF-
d8, 193K, arbitrary numbering scheme as shown in the insert): δ )
77.4 (C-1, d), 58.5 (C-4, t, JCH ) 133 Hz), 46.4 (C-5, q, JCH ) 132
Hz), 28.3 (C-3, t, JCH ) 150 Hz), 18.1 (broad, C-2, t, JCH ) 141 Hz).
Preparation of [CpFe(cod)] [Li(dme)] (4). A flame-dried three-
necked round-bottom flask equipped with a magnetic stirbar, a gas inlet,
a glass stopper and a stopcock connected to the vacuum line is charged
a Reagents and conditions: (a) complex 4 (10%), toluene (0.1 M), 90
°C, 80% (44), 63% (45); E ) COOEt.
has been intercepted in the coordination sphere of the 17e
Fe(+1) center (Scheme 14).30d It cannot be decided at this stage
whether 41 forms via an initial oxidative cyclization of two
butynes followed by insertion of the (coordinated) ethylene (path
A), or, conversely, implies an initial mixed oxidative cyclization
mode between butyne and ethene followed by insertion of the
second alkyne (path B); as the latter corresponds to the initial
steps of the Alder-ene manifold (e.g., 42 (R ) Me) f 43 f
44), we are inclined to favor this pathway (Scheme 15). It will
also explain why substrate 42 (R ) H) dimerizes to the [2+2+2]
cycloadduct 45, which is formed as a single isomer.
In any case, the rather unusual complexes 38 and 41
corroborate the competence of highly electron-richsneutral as
well as chargedsiron half-sandwich complexes in oxidative
cyclization processes and hence stimulate further investigations
into this still fairly unexplored terrain of organometallic
chemistry. The fact that even an Fe(+1) template with an odd
electron count is catalytically relevant and well behaved is a
hitherto largely unrecognized aspect that deserves particular
attention.
Conclusions. Ferrocene serves as a convenient and very
cheap starting material for the preparation of a family of low
valent iron olefin complexes of the formal oxidation states -2,
0 and +1. The structures of the ferrate complexes 1 and 4 in
9
J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008 2003

A R T I C L E S
Fu¨rstner et al.
under Ar with ferrocene (50.0 g, 268 mmol), freshly distilled 1,5-
cyclooctadiene (COD, 75 mL, 611 mmol) and DME (350 mL). The
resulting suspension is cooled to -50 °C before lithium sand (4.7 g,
680 mmol) is introduced. Stirring is continued at -20 °C for 4 h, during
which time a characteristic color change from orange to red indicates
the progress of the reaction. The resulting suspension, which hardly
settles, is allowed to stir for another hour at ambient temperature before
all insoluble materials are filtered off under Ar. The dark red filtrate is
stored at -30 °C for ca. 20 h, causing the formation of orange-red
crystals which are collected, washed with cold Et2O (2 × 100 mL),
and dried under vacuum (10-3 Torr) giving crude 4 as an orange-red
powder (85 g, 97%).
Analytically pure samples are obtained by recrystallization as
follows: COD (1 mL) is added to a suspension of a fraction of the
crude product (23.1 g) in toluene (100 mL) and the resulting mixture
is heated to reflux with stirring. Remaining undissolved materials are
filtered off while hot, and the resulting dark red, clear filtrate is kept
at 0 °C causing the precipitation of analytically pure 4 in form of dark
orange-red crystals which are filtered off and dried under vacuum (10-3
Torr) (11.6 g, 50%). The air- and moisture sensitive product can be
handled at ambient temperature without noticeable decomposition and
can be stored under Ar at -20 °C for extended periods of time (>1
year). 13C NMR (100 MHz, THF-d8, 193 K, arbitrary numbering scheme
as shown in the insert): δ ) 75.2 (C-1), 72.6 (C-5), 58.9 (C-4), 32.6
(C-3) ppm; The signal of the olefinic carbons is not detected under
these conditions; however, it is visible as a broad signal in the spectrum
of 4 recorded in toluene at 353 K: 13C NMR (100 MHz, toluene-d8):
δ ) 75.8 (C-1), 72.3 (C-5), 58.9 (C-4), 48.0 (br, C-2), 34.0 (C-3) ppm.
Preparation of Cp*Fe(tetraphenylcyclobutadiene) (38). Tolane
(367 mg, 2.06 mmol) is added to a red solution of complex 6 (246 mg,
0.995 mmol)32 in pentane (5 mL) at -78 °C. After stirring for 5 min,
the mixture is allowed to reach ambient temperature, during which time
a characteristic color change to dark brown and the evolution of gas is
noticed. After stirring for 1 h, the mustard-colored precipitate is filtered
off under Ar, the crude product is recrystallized from toluene, the
crystals are rinsed with cold (0 °C) pentane and dried in vacuo to give
complex 38 as a brown solid (420 mg, 77%). MS (EI): m/z (%): 548
(42), 547 (100) [M+], 412 (4), 369 (5), 133 (7); elemental analysis
calcd (%) for C38H35Fe (547.57): C 83.35, H 6.44; found: C 83.17, H
6.36.
Acknowledgment. We thank Prof. K. Jonas for valuable
discussions and advice, and Dr. R. Mynott for expert NMR
support. Generous financial support by the Fonds der Chemis-
chen Industrie, the Deutsch-Israelische Projektkooperation (DIP),
the Alexander-von-Humboldt Foundation (fellowship for R.M.)
and Pfizer Inc. (fellowship for K.M.) is gratefully acknowledged.
Supporting Information Available: Experimental part in-
cluding spectroscopic data of all new compounds. This material
is available free of charge via the Internet at http://pubs.acs.org.
JA0777180
9
2004 J. AM. CHEM. SOC. VOL. 130, NO. 6, 2008