Organolanthanide-catalyzed intra- and intermolecular tandem C-N and C-C bond-forming processes of aminodialkenes, aminodialkynes, aminoalkeneynes, and aminoalkynes. New regiospecific approaches to pyrrolizidine, indolizidine, pyrrole, and pyrazine skeleto

Article abstract of DOI:10.1021/ja972643t

This contribution describes catalytic tandem C-N and C-C bond-forming reactions involving the intramolecular hydroamination/bicyclization and intermolecular hydroamination/cyclization of olefins and alkynes using the organolanthanide complexes Cp'₂

Full text of DOI:10.1021/ja972643t

J. Am. Chem. Soc. 1998, 120, 1757-1771
1757
Organolanthanide-Catalyzed Intra- and Intermolecular Tandem C-N
and C-C Bond-Forming Processes of Aminodialkenes,
Aminodialkynes, Aminoalkeneynes, and Aminoalkynes. New
Regiospecific Approaches to Pyrrolizidine, Indolizidine, Pyrrole, and
Pyrazine Skeletons
Yanwu Li and Tobin J. Marks*
Contribution from the Department of Chemistry, Northwestern UniVersity, 2145 Sheridan Road,
EVanston, Illinois 60208-3113
ReceiVed August 1, 1997
Abstract: This contribution describes catalytic tandem C-N and C-C bond-forming reactions involving the
intramolecular hydroamination/bicyclization and intermolecular hydroamination/cyclization of olefins and
alkynes using the organolanthanide complexes Cp′2LnCH(SiMe3)2 and Me2SiCp′′2LnCH(SiMe3)2 (Cp′ ) η5-
Me5C5; Cp′′ ) η5-Me4C5; Ln ) lanthanide) as precatalysts. In the case of the intramolecular processes, substrates
of the structures RCtC(CH2)aNH(CH2)bCtCR, RCtC(CH2)cNH(CH2)dCHdCH2, and H2CdCH-(CH2)e-
NH(CH2)fCHdCH2 are regiospecifically bicyclized to the corresponding pyrrolizidine and indolizidine skeletons,
with turnover frequencies ranging from 2 to 777 h-1 at 21 °C and isolated product yields ranging from 85 to
93%. In the case of e ) 3 and f ) 1 mediated by Cp′2Sm-, the kinetic rate law is zero-order in substrate
concentration and first-order in lanthanide concentration. In the case of R ) Ph, c ) 3, and d ) 1, the
Cp′2Ln-catalyzed turnover frequencies fall precipitously with decreasing Ln+3 ionic radius. In the intermolecular
processes, substrates of the type HCtCCH2NHR undergo regiospecific coupling and cyclization to the
corresponding pyrroles MeCC(H)dC(CH2NHR)N(R)CH with high turnover frequencies where R and Nt
(h-1) are the following: CH2dCHCH2, 236 (60 °C); CH3CH2CH2, 208 (60 °C); CH2dCHCH2CH2CH2, 58
(60 °C). In addition, hydroamination/cyclization processes after intermolecular insertion can be effected when
R ) CH2dCHCH2, to afford a 2,7-dimethyldipyrrolo[1,2-a:1′,2′-d]pyrazine derivative via two successive
intramolecular olefin insertion processes. The mechanism for such tandem C-N and C-C bond formations
is postulated to involve turnover-limiting intra- or intermolecular alkene/alkyne insertion into the Ln-N
functionality, followed by rapid intramolecular insertion of a pendant CdC/CtC-containing functionality into
the resulting Ln-C bond (prior to protonolysis). Such a scenario is consistent with well-documented, stepwise
transformations in organo-f-element-catalyzed insertions of unsaturated carbon-carbon multiple bonds into
metal-amide and metal-alkyl functionalities.
Introduction
carbon-carbon functionalities (hydroamination) can likewise
be catalyzed by a variety of metal complexes.1,2,5-7 The
Carbon-nitrogen and carbon-carbon bond-forming reactions
are important fundamental transformations in synthetic chem-
istry. Such reactions are in general most useful and efficient
when performed catalytically, rather than stoichiometrically.1,2
Catalytic carbon-carbon bond-forming transformations medi-
ated by early and late transition metal catalysts have been studied
extensively.1,3,4 The addition of N-H bonds to unsaturated
(2) For leading reviews of amine additions to olefins and alkynes, see:
(a) Hegedus, L. S. Angew. Chem., Int. Ed. Engl. 1988, 27, 1113-1126. (b)
Gasc, M. B.; Latties, A.; Perie, J. J. Tetrahedron 1983, 39, 703-731. (c)
Ba¨ckvall, J.-E. Acc. Chem. Res. 1983, 16, 335-342. (d) Ja¨ger, V.; Viehe,
H. G. Houben-Weyl, Methoden der Organischen Chemie; Thieme Verlag:
Stuttgart, Germany, 1977; Vol. 5/2a, pp 713-724. (e) Suminov, S. I.; Kost,
A. N. Russ. Chem. ReV. 1969, 38, 884-899. (f) Chekulaeva, I. A.;
Kondrat’eva, L. V. Russ. Chem. ReV. 1965, 34, 669-680.
(3) Alkene insertions involving Ln-alkyl bonds: (a) Li, Y.; Marks, T.
J. J. Am. Chem. Soc. 1996, 118, 707-708 (preliminary communication of
some aspects of this work). (b) Molander, G. A.; Nichols, P. J. J. Am. Chem.
Soc. 1995, 117, 4415-4416. (c) Casey, C. P.; Hallenbeck, S. L.; Pollok,
D. W.; Landis, C. R. J. Am. Chem. Soc. 1995, 117, 9770-9771. (d) Fu,
P.-F.; Brard, L.; Li, Y.; Marks, T. J. J. Am. Chem. Soc. 1995, 117, 7157-
7168. (e) Fu, P.-F.; Marks, T. J. J. Am. Chem. Soc. 1995, 117, 10747-
10748. (f) Schaverien, C. J. AdV. Organomet. Chem. 1994, 36, 283-362.
(g) Yang, X.; Seyam, A. M.; Fu, P.-F.; Marks, T. J. Macromolecules 1994,
27, 4625-4626. (h) Molander, G. A.; Hoberg, J. O. J. Am. Chem. Soc.
1992, 114, 3123-3124. (i) den Haan, K. H.; de Boer, J. L.; Teuben, J. H.;
Spek, A. L.; Kajic-Prodic, B.; Hays, G. R.; Huis, R. Organometallics 1986,
5, 1726-1733. (j) Watson, P. L.; Parshall, G. W. Acc. Chem. Res. 1985,
18, 51-55. (k) Jeske, G.; Lauke, H.; Mauermann, H.; Swepston, P. N.;
Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8091-8103.
(1) For general reviews, see: (a) Larock, R. C. In AdVances in Metal-
Organic Chemistry; JAI: Greenwich, U.K., 1994; Vol. 3, pp 97-224. (b)
Parshall, G. W.; Ittel, S. D. Homogeneous Catalysis; Wiley-Interscience:
New York, 1992. (c) March, J. AdVanced Organic Chemistry, 4th ed.; J.
Wiley & Sons: New York, 1992; Chapter 15, pp 768-770, pp 790-812,
and references therein. (d) Brunet, J.-J.; Neibecker, D.; Niedercorn, F. J.
Mol. Catal. 1989, 49, 235-259. (e) Collman, J. P.; Hegedus, L. S.; Norton,
J. R.; Finke, R. G. Principles and Applications of Organotransition Metal
Chemistry; University Science Books: Mill Valley, CA, 1987; Chapters
7.4 and 17.1. (f) Trost, B. M.; Verhoeven, T. R. In ComprehensiVe
Organometallic Chemistry; Wilkinson, G., Stone, F. G. A., Abel, E. W.,
Eds.; Pergamon Press: Oxford, U.K., 1982; Vol. 8, pp 892-895 and
references therein. (g) Gibson, M. S. In The Chemistry of the Amino Group;
Patai, S., Ed.; Interscience: New York, 1968; pp 61-65.
S0002-7863(97)02643-7 CCC: $15.00 © 1998 American Chemical Society
Published on Web 03/04/1998

1758 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Li and Marks
Scheme 1. Possible Scenarios for Organolanthanide-
feasibility of such transformations therefore raises the interesting
question of whether individual C-N and C-C bond-forming
steps can be coupled in sequence to assemble more elaborate
polycyclic, heteroatom-containing skeletons (e.g., pyrrole, pyr-
rolizidine, indolizidine, pyrazine, and other alkaloid frame-
works)8-10 using a single metal center in a single catalytic cycle.
In this regard, lanthanide metal centers11 exhibit a number of
distinctive as well as potentially informative and useful char-
acteristics for the activation not only of unsaturated carbon-
carbon multiple bonds but of amine groups as well. The unique
characteristics of lanthanide ions include high electrophilicity,
very large ionic radii (resulting in high coordination numbers
Catalyzed Aminoalkenyne Hydroamination/Cyclization
(4) Alkyne insertions involving Ln-alkyl or An-alkyl bonds: (a) Straub,
T.; Haskel, A.; Eisen, M. S. J. Am. Chem. Soc. 1995, 117, 6364-6365. (b)
Duchateau, R.; Van Wee. C. T.; Meetsma, A.; Teuben, J. H. J. Am.. Chem.
Soc. 1993, 115, 4931-4932. (c) Heeres, H. J.; Nijhoff, J.; Teuben, J. H.
Organometallics 1993, 12, 2609-2617. (d) Evans, W. J.; Keyer, R. A.;
Ziller, J. W. Organometallics 1993, 12, 2618-2633. (e) Heeres, H. J.;
Teuben, J. H. Organometallics 1991, 10, 1980-1986 and references therein.
(f) Heeres, H. J.; Heeres, A.; Teuben, J. H. Organometallics 1990, 9, 1508-
1510. (g) Heeres, H. J.; Meetsma, A.; Teuben, J. H.; Rogers, R. D.
Organometallics 1989, 8, 2637-2646. (h) Den Haan, K. H.; Wielstra, Y.;
Teuben, J. H. Organometallics 1987, 6, 2053-2060. (i) Reference 3a.
(5) (a) McGrane, P. L.; Livinghouse, T. J. Am. Chem. Soc. 1993, 115,
11485-11489. (b) McGrane, P. L.; Livinghouse, T. J. Org. Chem. 1992,
57, 1323-1324. (c) McGrane, P. L.; Jensen, M.; Livinghouse, T. J. Am.
Chem. Soc. 1992, 114, 5459-5460. (d) Walsh, P. J.; Baranger, A. M.;
Bergman, R. G. J. Am. Chem. Soc. 1992, 114, 1708-1719. This article
begins with a review of conventional hydroamination processes involving
unsaturated carbon-carbon bonds. (e) Tamaru, Y.; Hojo, M.; Higashimura,
H.; Yoshida, Z.-I. J. Am. Chem. Soc. 1988, 110, 3994-4002 and references
therein. The authors summarize the strengths and weaknesses of platinum
metal-catalyzed transformations. (f) Fukuda, Y.; Utimoto, K.; Nozaki, H.
Heterocycles 1987, 25, 297-300. (g) Pugin, B.; Venanzi, L. M. J.
Organomet. Chem. 1981, 214, 125-133, and references therein. (h)
Reference 1f,g.
and coordinative unsaturation), relatively constrained/immobile
yet tunable ancillary ligation, and facile bond activation via
concerted four-centered σ bond metathesis processes rather than
by conventional two-electron oxidative addition/reductive elimi-
nation sequences.
The facility of organolanthanide-catalyzed hydroamination/
cyclization via olefin or alkyne insertion into the Ln-N bonds
followed by rapid protonolysis (e.g., eqs 1 and 2; Cp′ ) η5-
Me5C5; Ln ) lanthanide) has been recently documented and
mechanistically characterized (e.g., Scheme 1, steps i and ii).6,7
It would therefore be of interest to determine if an alternative
process (Scheme 1, step iii) to protonolysis (Scheme 1, step ii)
could be coupled with metal-alkyl/alkenyl bond formation
(Scheme 1, step i) to effect additional bond-forming processes
(Scheme 1, steps i + iii).
Organolanthanide complexes of the type Cp′2LnR (R ) H,
CH(SiMe3)2) are highly efficient catalysts for olefin insertion
processes, exhibiting, for example, turnover frequencies in
excess of 1500 s-1 at 25 °C and 1 atm of pressure for ethylene
polymerization.3j,12 Similar rapid processes involving intramo-
lecular olefin insertions into Ln-C bonds might conceivably
(6) Alkene insertions involving Ln-heteroatom bonds: (a) Li, Y.; Marks,
T. J. Organometallics 1996, 15, 3770-3772. (b) Giardello, M. A.;
Conticello, V. P.; Brard, L.; Gagne´, M. R.; Marks, T. J. J. Am. Chem. Soc.
1994, 116, 10241-10254. (c) Gagne´, M. R.; Stern, C. L.; Marks T. J. J.
Am. Chem. Soc. 1992, 114, 275-294. (d) Giardello, M. A.; King, W. A.;
Nolan, S. P.; Porchia M.; Sishta, C.; Marks, T. J. In Energetics of
Organometallic Species; Martinho Simoes, J. A., Ed.; Kluwer: Dordrecht,
The Netherlands, 1992; pp 35-54. (e) Gagne´, M. R.; Brard, L.; Conticello,
V. P.; Giardello, M. A.; Stern, C. L.; Marks, T. J. Organometallics 1992,
11, 2003-2005. (f) Gagne´, M. R.; Nolan, S. P.; Marks, T. J. Organo-
metallics 1990, 9, 1716-1718. (g) Reference 3a.
(7) Alkyne insertions involving Ln-heteroatom bonds: (a) Li, Y.; Marks,
T. J. J. Am. Chem. Soc. 1996, 118, 9295-9306. (b) Li, Y.; Fu, P.-F.; Marks,
T. J. Organometallics 1994, 13, 439-440.
(8) For some leading reviews of pyrrolizidine and indolizidine chemistry,
see: (a) Robins, D. J. Nat. Prod. Rep. 1994, 11, 613-619 and references
therein. (b) Michael, J. P. Nat. Prod. Rep. 1994, 11, 639-657 and references
therein. (c) Daly, J. W.; Spande, T. F. In Alkaloids: Chemical and Biological
PerspectiVes; Pelletier, S. W., Ed.; Wiley: New York, 1986; Vol. 4, Chapter
1. (d) Mattocks, A. R. Chemistry and Toxicology of Pyrrolizidine Alkaloids;
Academic Press: Orlando, FL, 1986. (e) Robins, D. J. AdV. Heterocycl.
Chem. 1979, 24, 247-291.
(9) For some leading reviews of pyrrole and pyrazine chemistry, see:
(a) Plunkett, A. O. Nat. Prod. Rep. 1994, 11, 581-590 and references
therein. (b) Pinder, A. R. Nat. Prod. Rep. 1992, 9, 491-504 and references
therein. (c) Toube, T. P. In Pyrroles; Jones, R. A., Ed.; Wiley: New York,
1992; Part 2, Chapter 1 and references therein. (d) Cirrincione, G.; Almerico,
A. M.; Aiello, E. In Pyrroles; Jones, R. A., Ed.; Wiley: New York, 1992;
Part 2, Chapter 3 and references therein. (e) Barlin, G. B. The Pyrazines;
John Wiley & Sons: New York, 1982.
(10) For representative synthetic approaches, see: (a) Hudlicky, T.; Reed,
J. W. In ComprehensiVe Organic Synthesis; Trost, B. M., Fleming, I., Eds.;
Pergamon Press: Oxford, U.K., 1991; Vol. 5, pp 937-965 and references
therein. (b) Hassner, A.; Rai, K. M. L. In ref 10a, Vol. 1, pp 557-559 and
references therein. (c) Shiosaki, K. In ref 10a, Vol. 2, pp 882-885 and
references therein. (d) Overman, L. E.; Ricca, D. J. In ref 10a, Vol. 2, pp
1040-1044 and references therein. (e) Meth-Cohn, O. In ref 10a, Vol. 2,
pp 789-790 and references therein. (f) Sweger, R. W.; Czarnik, A. W. In
ref 10a, Vol. 5, pp 581-583 and references therein. (g) Bumagin, N. A.;
Nikitina, A. F.; Beletskaya, I. P. Russ. J. Org. Chem. 1994, 30 (10), 1619-
1629 (Engl. Transl.). (h) Reference 5b and references therein. (i) Jefford,
C. W.; Tang, Q.; Zaslona, A. J. Am. Chem. Soc. 1991, 113, 3513-
3518.
(11) For general organolanthanide references, see: (a) Edelmann, F. T.
In ComprehensiVe Organometallic Chemistry II; Abel, E. W., Stone, F. G.
A., Wilkinson, G., Eds.; Elsevier: Oxford, U.K., 1995; Vol. 4, pp 11-812
and references therein. (b) Edelmann, F. T. Angew Chem., Int. Ed. Engl.
1995, 34, 2466-2488. (c) Schumann, H.; Meese-Marktscheffel, J. A.; Esser,
L. Chem. ReV. 1995, 95, 865-986 and references therein. (d) Schaverien,
C. J. AdV. Organomet. Chem. 1994, 36, 283-362 and references therein.
(e) Evans, W. J. AdV. Organomet. Chem. 1985, 24, 131-177 and references
therein. (f) Schumann, H. In Fundamental and Technological Aspects of
Organo-f-Element Chemistry; Marks, T. J., Fragala I., Eds.; D. Reidel:
Dordrecht, Holland, 1985; Chapter 1. (g) Kagan, H. B.; Namy, J. L. In
Handbook on the Physics and Chemistry of Rare Earths; Gschneider, K.
A., Eyring, L., Eds.; Elvesier: Amsterdam, 1984; Chapter 50. (h) Forsberg,
J. H.; Hoeller, T. In Gmelin Handbook of Inorganic Chemistry; Moeller,
T., Kmerke, Y., Schleitzer-Rust, E., Eds.; Springer: Berlin, 1983; pp 137-
303.

Catalytic Tandem C-N and C-C Bond-Forming Reactions
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1759
be competitive with Ln-C protonolysis in Scheme 1 (step ii).
To test this hypothesis, a variety of bisunsaturated secondary
amines were synthesized and investigated in catalytic reactions
designed to facilitate tandem and sequential C-N and C-C
bond-forming transformations.3a
hydroamination/cyclization within organolanthanide coordina-
tion spheres, including an investigation of reaction scope and
metal ion size effects as well as kinetics and mechanism, is
presented.
Although the direct intermolecular addition of amines to
olefins or alkynes is a priori an attractive transformation in
organic synthesis, such simple hydroamination processes have
not generally proven to be efficient and frequently lack
generality as well as regioselectivity.2,5e The catalysts for these
simple additions include alkali metals,2,13 solid acids,14 and late
transition metals15 which activate either the amine or carbon-
carbon multiple bond functionality. However, efficient homo-
geneous processes are rare.5d,6a,15,16 Of the effective intermo-
lecular hydroamination processes, organolanthanide-mediated
transformations offer some promise with significant turnover
frequencies compared to most transition metal-catalyzed proc-
esses (eqs 1 and 2). Despite this, there remain problems
Experimental Section
Materials and Methods. All manipulations of air-sensitive materi-
als were carried out with rigorous exclusion of oxygen and moisture
using methodologies described previously.6,7 Argon (Matheson, pre-
purified) was purified by passage through a MnO oxygen-removal
column17 and a Davison 4 A molecular sieve column. Before use, all
solvents were distilled under dry nitrogen over appropriate drying agents
(sodium benzophenone ketyl, metal hydrides, or Na/K alloy, except
for chlorinated solvents). Deuterium oxide and chloroform-d were
purchased from Cambridge Isotope Laboratories. Benzene-d6 and
toluene-d8 (Cambridge Isotope Laboratories, all 99+ atom % D), used
for NMR reactions and kinetic measurements, were stored in vacuo
over Na/K alloy in resealable bulbs and were vacuum-transferred
immediately prior to use. All organic starting materials were purchased
from Aldrich Chemical Co., Farchan Laboratories Inc., or Lancaster
Synthesis Inc. and, when appropriate, were distilled prior to use. For
intramolecular hydroamination/bicyclization reactions, the substrates
N-allyl-4-pentyn-1-amine7b (5), N-allyl-5-(trimethylsilyl)-4-pentyn-1-
amine7b (7), N-4-penten-5′-(trimethylsilyl)-4′-pentyn-1-amine7b (9), N-4-
pentyn-4′-penten-1-amine7b (11), and N-allyl-4-penten-1-amine18 (15)
were synthesized according to literature procedures. The new com-
pounds N-allyl-5-phenyl-4-pentyn-1-amine (1), N-allyl-4-hexyn-1-amine
(3), N-2-butyn-4′-hexyn-1-amine (13), N-allyl-5-hexen-1-amine (17),
N-2-butyn-4′-penten-1-amine (19), N-3-(trimethylsilyl)-2-propyn-4′-
penten-1-amine (21), and N-4-(trimethylsilyl)-3-butyn-4′-penten-1-
amine (23) were synthesized via modifications of literature methods
as described below. Substrates 1, 3, 7, 9, 13, 15, 17, 19, 21, and 23
were dried by stirring over CaH2, and substrates 5 and 11 were dried
by stirring over BaO. All were then additionally dried by repeated
vacuum transfer onto and from, freshly activated Davison 4 A molecular
sieves, were degassed by freeze-pump-thaw cycles, and were finally
stored in vacuum-tight containers. For intermolecular hydroamination/
cyclization reactions, N-allyl-N-propargylamine (25), N-n-propyl-N-
propargylamine (27), and N-4-pentenyl-N-propargylamine (29) were
synthesized via modifications of literature methods as described below
and were dried in a manner analogous to 5 and 11 above. The
organolanthanide precatalysts Cp′2LnCH(SiMe3)2 (Ln ) La, Nd, Sm,
Lu; Cp′ ) η5-Me5C5)12a and Me2SiCp′′2LnCH(SiMe3)2 (Cp′′ ) η5-
Me4C5)12b were prepared by published procedures.
associated with limited reaction rates since a large excess of
more nucleophilic amine substrates inhibits olefin or alkyne
interactions with the strongly Lewis acidic metal centers, thereby
retarding carbon-carbon multiple-bond insertions. An attractive
alternative approach would be to employ olefin and alkyne
substrates also containing a pendant Lewis basic group, such
as an amine moiety, to temporarily anchor the substrate at the
metal center. This would be expected to facilitate activation
of the unsaturated carbon-carbon multiple bond by proximity
within the lanthanide coordinative sphere. There is substantial
precedent for L2LnNRR′(HNRR′) adducts in organolanthanide
chemistry (L ) cyclopentadienyl-type ligand).6b,c Furthermore,
if this approach were coupled with a subsequent rapid insertive
C-C bond-forming process, the resulting two reactions (inter-
molecular C-N bond and intramolecular C-C bond formation)
would afford pyrrole-type heterocycles in the same catalytic
reaction and provide a new type of tandem inter- + intramo-
lecular coupling reaction.
Physical and Analytical Measurements. NMR spectra were
recorded on either a Varian VXRS 300 (FT, 300 MHz (1H), 75 MHz
(13C)) or UNITYplus 400 (FT, 400 MHz (1H), 100 MHz (13C))
instrument. Chemical shifts (δ) for 1H and 13C are referenced to internal
solvent resonances and reported relative to SiMe4. NMR experiments
on air-sensitive samples were conducted in Teflon valve-sealed tubes
(J. Young). Analytical gas chromatography was performed on a Varian
Model 3700 gas chromatograph with FID detection and a Hewlett-
Packard 3390A digital recorder/integrator using a 0.125 in. i.d. column
with 3.8% w/w SE-30 liquid phase on Chromosorb W support. GC/
MS studies were conducted on a VG 70-250 SE instrument with 70-
eV electron impact ionization. IR spectra were recorded using a Nicolet
520 FT-IR spectrometer with an MCT detector. Melting and boiling
points are uncorrected.
In this contribution, we present the first investigation of
tandem C-N and C-C bond-forming processes catalyzed by
organolanthanides.3a A full discussion of efficient, regiospecific
intramolecular hydroamination/bicyclization and intermolecular
(12) (a) Jeske, G.; Schock, L. E.; Mauermann, H.; Swepston, P. N.;
Schumann, H.; Marks, T. J. J. Am. Chem. Soc. 1985, 107, 8103-8110. (b)
Jeske, G.; Lauke, H.; Mauermann, H.; Schumann, H.; Marks, T. J. J. Am.
Chem. Soc. 1985, 107, 8111-8118. (c) Reference 3k. (d) Mauermann, H.;
Marks, T. J. Organometallics 1985, 4, 200-202.
(13) For catalytic intermolecular hydroamination of olefins using alkali
metals, see: (a) Pez, G. P.; Galle, J. E. Pure Appl. Chem. 1985, 57, 1917-
1926. (b) Howk, B. W.; Little, E. L.; Scott, S. L.; Whitman, G. M. J. Am.
Chem. Soc. 1954, 76, 1899-1902.
(14) (a) Deeba, M.; Ford, M. E. J. Org. Chem. 1988, 53, 4594-4596.
(b) Deeba, M.; Ford, M. E.; Johnson, T. A. J. Chem. Soc., Chem. Commun.
1987, 562-563.
Synthesis of N-Allyl-5-phenyl-4-pentyn-1-amine (1). A solution
of 50.0 g (0.477 mol) of phenylacetylene in 200 mL of THF was treated
with 0.477 mol of n-BuLi at 0 °C over a period of 2 h, warmed to
room temperature, and stirred for 2 h. After the resulting solution was
cooled to -78 °C, 110.0 g (0.534 mol) of BrCH2CH2CH2Br was rapidly
(15) For catalytic intermolecular hydroamination of alkenes and alkynes
using transition metals, see: (a) Dorta, R.; Egli, P.; Zu¨rcher, F.; Togni A.
J. Am. Chem. Soc. 1997, 119, 10857-10858. (b) Brunet, J.-J.; Neibecker,
D.; Philippot, K. Tetrahedron Lett. 1993, 34, 3877-3880. (c) Casalnuovo,
A.L.; Calabrese, J. C.; Milstein, D. J. Am. Chem. Soc. 1988, 110, 6738-
6744. (d) Barluenga, J.; Azar, F.; Liz, R.; Rodes, R. J. Chem. Soc., Perkin
Trans. 1 1980, 2732-2737. (e) Coulson, D. R. Tetrahedron Lett. 1971,
429-430. Coulson, D. R. U.S. Patent 3758586, 1973.
(17) (a) He, M.-Y.; Xiong, G.; Toscano, P. J.; Burwell, R. L., Jr.; Marks,
T. J. J. Am. Chem. Soc. 1985, 107, 641-652. (b) Moeseler, R.; Horvath,
B.; Lindenau, D.; Horvath, E. G.; Krauss, H. L. Z. Naturforsch. 1976, 31B,
892-893. (c) McIlwrick, C. R.; Phillips, C. S. G. J. Phys. E 1973, 6, 1208-
1210.
(18) Newcomb, M.; Marquardt, D. J.; Deeb, T. M. Tetrahedron 1990,
46, 2329-2344.
(16) Haskel, A. H.; Straub, T.; Eisen, M. S. Private communication.

1760 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Li and Marks
syringed into the stirring solution and the mixture was stirred for 2 h
at -78 °C, warmed to room temperature for 2 h, and then refluxed
overnight. Next, 150 mL of water was poured into the cooled mixture,
the organic phase was separated off, and the aqueous phase was
extracted with diethyl ether (3 × 100 mL). The combined organic
phase and diethyl ether extracts were dried over MgSO4 and filtered,
and the solvents were removed by rotary evaporation. Distillation
(111-114 °C/0.17 Torr) of the residue gave 41.7 g (39% yield) of
PhCtCCH2CH2CH2Br. A solution of 19.0 g (0.085 mol) of 1-bromo-
5-phenyl-4-pentyne and 34.0 g (0.590 mol) of allylamine in a sealed
150-mL storage tube was heated at 60 °C for 4 days. The reaction
mixture was then poured into 250 mL of water, the organic layer was
separated, and the aqueous phase was extracted with diethyl ether (3
× 50 mL). The combined organic phase and ether solution were dried
over MgSO4 and filtered, and the ether and excess allylamine were
removed by rotary evaporation. Distillation (127-129 °C/0.15 Torr)
gave 1 as a colorless oil (14.0 g, 83% yield).
1H NMR (300 MHz, C6D6): δ 7.45 (m, 2H, Ph), 6.94 (m, 3H, Ph),
5.79 (m, 1H, CHd), 5.11-4.93 (m, 2H, CH2d), 2.98 (d, J ) 6.0 Hz,
2H, NCH2-CHd), 2.47 (t, J ) 6.8 Hz, 2H, CH2N), 2.29 (t, J ) 7.1
Hz, 2H, tC-CH2), 1.52 (m, 2H, CH2), 0.49 (br, 1H, NH). 13C (75
MHz, C6D6): δ 137.6, 131.5, 128.1, 127.3, 124.4, 114.6, 90.2, 81.1,
52.2, 48.0, 29.2, 17.1. MS (relative abundance): M+ (45), M+ - 1
(86), M+ + 1 (7), 184.1 (62), 172.1 (57), 156.0 (22), 141.0 (34), 128.0
(37), 115.0 (54), 91.0 (14), 82.0 (13), 70.0 (100), 56.0 (11), 41.0 (68).
HRMS: Calcd for C14H17N: 199.1361. Found: 199.1351.
Synthesis of N-Allyl-4-hexyn-1-amine (3). A solution of 6.6 g
(0.056 mol) of 1-chloro-4-hexyne and 50.0 g (0.86 mol) of allylamine
was heated in a sealed tube at 60 °C for 1 week. After the solution
was cooled to room temperature and poured into a separatory funnel
containing 200 mL of diethyl ether and 30 mL of H2O, the aqueous
phase was separated and extracted with 3 × 100-mL portions of diethyl
ether. The combined ethereal extracts and the organic phase were
washed with 70 mL of brine, dried over MgSO4, and concentrated by
rotary evaporation. Distillation (108-110 °C) gave 3 as a colorless
oil (6.0 g, 77% yield).
1H NMR (300 MHz, CDCl3): δ 5.89 (m, 1H, CHd), 5.10 (m, 2H,
CH2d), 3.32 (d, J ) 6.0 Hz, 2H, NCH2CHd), 2.68 (t, J ) 7.1 Hz, 2H,
NCH2), 2.16 (m, 2H, CH2), 1.47 (br, 1H, NH). 13C NMR (75 MHz,
CDCl3): δ 136.8, 115.8, 78.6, 75.8, 52.3, 48.4, 29.1, 16.6, 3.45. MS
(relative abundance): M+ (25), M+ + 1 (3), 122.1 (69), 110.1 (33),
94.1 (19), 79.1 (12), 70.1 (100), 53.1 (16), 41.1 (81). HRMS: Calcd
for C9H14N (M+ - H): 136.1126. Found: 136.1122.
1-amine under Ar at room temperature. The reaction mixture was then
heated in an oil bath at 55 °C for 5 h until the starting material was
consumed (TLC monitoring) and then allowed to cool to 25 °C. The
reaction mixture was poured into a separatory funnel containing 500
mL of aqueous NaOH solution (1%). The resulting solution was
extracted with diethyl ether (3 × 200 mL). The combined ether extracts
were washed with brine (50 mL), dried over MgSO4, filtered,
concentrated with a rotary evaporator, and purified by flash chroma-
tography (hexane:diethyl ether ) 1:1). After distillation (126-128 °C/
19 Torr), 7.4 g (81% yield) of CH3CtCCH2CH2CH2NHCH2CtCCH3
was obtained as a colorless oil.
1H NMR (300 MHz, C6D6): δ 3.19 (m, 2H, NCH2CtC), 2.61 (t, J
) 6.9 Hz, 2H, NCH2), 2.11 (m, 2H, CH2CtC), 1.54-144 (m, 8H,
CH2, 2CH3), 0.53 (br, 1H, NH). 13C NMR (75 MHz, C6D6): δ 79.3,
78.4, 78.2, 75.6, 47.9, 38.8, 29.7, 16.9, 3.4, 3.3. MS (relative
abundance): M+ (12), M+ - 1 (35), M+ - 2 (3), 134.1 (100), 120.1
(19), 106.1 (17), 94.1 (10), 82.1 (48), 68.1 (7), 53.1 (52), 42.0 (16).
HRMS: Calcd for C10H14N (M+ - H): 148.1126. Found: 148.1110.
Synthesis of N-Allyl-4-penten-1-amine (15). A mixture of 25.0 g
(0.168 mol) of 5-bromo-1-pentene and 38.0 g (0.64 mol) of allylamine
in a sealed 150-mL storage tube was heated at 60 °C for 4 days. After
the reaction mixture was poured into 250 mL of water, the organic
layer was separated off and the aqueous phase washed with diethyl
ether (3 × 50 mL). The combined organic phase and ether solution
was dried over MgSO4 and filtered, and the ether and excess allylamine
were removed by rotary evaporation. Distillation (70-71 °C/25 Torr;
lit.18 bp 149-150 °C) of the residue gave 36.0 g (72% yield) of
CH2dCHCH2CH2CH2NHCH2CHdCH2 as a colorless oil. The NMR
data agree well with published data.18
1H NMR (400 MHz, CDCl3): δ 5.95 (m, 1H, NCH2CHd), 5.85
(m, 1H, CHd), 5.14 (m, 2H, NCH2CHdCH2), 5.01 (m, 2H, CH2d),
3.28 (d, J ) 6.0 Hz, 2H, NCH2CHd), 2.66 (t, J ) 7.2 Hz, 2H, NCH2),
2.13 (m, 2H, dCH-CH2), 1.63 (m, 2H, CH2), 1.05 (br, 1H, NH). 13
C
NMR (100 MHz, CDCl3): δ 138.0, 136.7, 115.2, 114.2, 52.1, 48.5,
31.2, 28.9. MS (relative abundance): M+ - 1 (3), 110.1 (3), 97.1 (5),
84.1 (9), 82.1 (10), 70.1 (100), 65.1 (8), 42.0 (9). HRMS: Calcd for
C8H14N (M+ - H): 124.1126. Found: 124.1117.
Synthesis of N-Allyl-5-hexen-1-amine (17). A procedure similar
to that for 15 was used for N-allyl-5-hexenamine via reaction of
6-bromo-1-hexene (4.9 g, 0.030 mol) with allylamine (38.0 g, 0.570
mol). Distillation (119-121 °C/134 Torr) of the residue gave 3.0 g
(72% yield) of CH2dCHCH2CH2CH2CH2NHCH2CHdCH2 as a color-
less oil.
Preparation of N-2-Butynyl-4′-hexyn-1-amine (13). (a) 2-Butyn-
1-amine. A suspension of 79.0 g (1.20 mol) of NaN3 in 200 mL of
DMF was treated dropwise at 25 °C for 30 min with 21.4 g (0.024
mol) of 2-butynyl 1-chloride, which was generated from 2-butynol (49.0
g, 0.699 mol), thionyl chloride (90.0 g, 0.756 mol), and 0.3 mL of
pyridine in 250 mL of boiling anhydrous diethyl ether in the usual
manner.19 The mixture was stirred overnight, the precipitate removed
by filtration, and the filtrate poured into a separatory funnel containing
800 mL of diethyl ether. The ether layer was washed with water (200
mL) and brine (100 mL), dried over MgSO4, filtered, and concentrated
by rotary evaporation to afford CH3CtCCH2N3, which was used
immediately in the next step. 1H NMR (300 MHz, CDCl3): δ 3.88 (s,
2H, CH2), 1.90 (s, 3H, CH3). The CH3CtCCH2N3 in 150 mL of
anhydrous diethyl ether was treated dropwise with LAH (0.177 mol, 4
equiv) in 50 mL of anhydrous diethyl ether at 0 °C for 30 min and
then stirred for 30 min. The resulting mixture was quenched by careful
addition of wet diethyl ether (ca. 200 mL) and a few drops of H2O.
After separation, the organic phase was washed with 200 mL of brine,
dried over MgSO4, filtered, concentrated with a rotary evaporator, and
distilled at 110-112 °C to give 9.1 g (55% yield) of the title compound.
1H NMR (300 MHz, CDCl3): δ 5.87 (m, 1H, NCH2CHd), 5.77
(m, 1H, CHd), 5.11 (m, 2H, NCH2CHdCH2), 4.94 (m, 2H, CH2d),
3.22 (d, J ) 6.9 Hz, 2H, NCH2CHd), 2.59 (t, J ) 7.1 Hz, 2H, NCH2),
2.04 (m, 2H, CH2-CHd), 1.59 (br, 1H, NH), 1.49 (m, 2H, CH2CH2N),
1.39 (m, 2H, CH2CH2CH2N). 13C NMR (75 MHz, CDCl3): δ 138.7,
136.7, 115.9, 114.5, 52.4, 49.2, 33.6, 29.5, 26.6. MS (relative
abundance): M+ (7), M+ + 1 (1), 110.1 (5), 96.1 (15), 84.1 (7), 70.0
(100), 56.0 (12), 41.0 (57). HRMS: Calcd for C9H16N (M+ - H):
138.1283. Found: 138.1285.
Preparation of N-2-Butynyl-4′-penten-1-amine (19). (a) 4-Penten-
1-yl Methanesulfonate. A mixture of 9.9 g (0.12 mol) of 4-penten-
1-ol and 26 mL of Et3N in 300 mL of CH2Cl2 was cooled to ca. -5
°C, and 9.7 mL (0.13 mol) of methanesulfonyl chloride was slowly
added over a period of 10 min. The solution was then stirred at ca.
-5 °C for 40 min, and 40 mL of ice water was added to quench the
reaction. The organic phase was separated and washed successively
with 50 mL of 1 M HCl, 50 mL of saturated Na2CO3 solution, and 50
mL of brine and was then dried over MgSO4. The solvent was removed
from the filtrate by rotary evaporation to afford 18.5 g (98% yield) of
CH2dCHCH2CH2CH2OSO2Me as a coloreless oil.
5
1H NMR (300 MHz, CDCl3): δ 3.35 (q, J ) 1.8 Hz, 2H, NCH2),
5
1H NMR (300 MHz, CDCl3): δ 5.76 (m, 1H, CHd), 5.03 (m, 2H,
CH2d), 4.22 (t, J ) 6.5 Hz, 2H, CH2O), 2.99 (s, 3H, Me), 2.16 (m,
2H, CH2CHd), 1.84 (m, 2H, CH2).
1.78 (t, J ) 1.8 Hz, 3H, CH3), 1.39 (br, 2H, NH2).
(b) N-2-Butynyl-4′-hexyn-1-amine. To a solution of 12.0 g (0.068
mol) of crude 4-hexyn-1-yl methanesulfonate19 and 200 mg of NaI in
40 mL of anhydrous DMSO was added 9.5 g (0.138 mol) of 2-butynyl-
(b) N-2-Butynyl-4′-penten-1-amine. Using a method similar to that
for N-2-butynyl-4′-hexyn-1-amine (9), the title compound was prepared
using 6.0 g (0.087 mol) of 2-butyn-1-amine and 7.0 g (0.043 mol) of
4-penten-1-yl methanesulfonate in 45 mL of anhydrous DMSO. By
(19) Arnold, H.; Overman, L. E.; Sharp, M. J.; Witschel, M. C. In
Organic Syntheses; Meyers, A. I., Ed.; John Wiley & Sons Inc.: New York,
1992; Vol. 70, pp 111-119.

Catalytic Tandem C-N and C-C Bond-Forming Reactions
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1761
following the procedure above, distillation (101-104 °C/25 Torr) gave
19 as a colorless oil (3.6 g, 62% yield).
98.1 (100), 73.1 (19), 44.1 (39), 41.1 (15). HRMS: Calcd for C12H23-
NSi: 209.1600. Found: 209.1602.
1H NMR (300 MHz, CDCl3): δ 5.80 (m, 1H, CHd), 4.97 (m, 2H,
Synthesis of N-Allylpropargylamine (25). A mixture of 7.0 g
(0.092 mol) of propargyl chloride and 38.1 g (0.666 mol) of allylamine
in a sealed 150-mL storage tube was heated at 60 °C for 2 days. The
reaction mixture was then poured into a separatory funnel containing
100 mL of water and 200 mL of diethyl ether, the organic layer was
separated off, and the aqueous phase was extracted with 3 × 150 mL
of diethyl ether. The combined organic phase and ether solution was
then dried over MgSO4 and filtered, and the ether and excess allylamine
were removed by rotary evaporation. Distillation (119-121 °C) of
the residue gave 5.8 g (66% yield) of CH2dCHCH2NHCH2CtCH as
a colorless oil.
5
CH2d), 3.35 (q, J ) 2.3 Hz, 2H, NCH2CtC), 2.66 (t, J ) 7.2 Hz,
5
2H, CH2N), 2.08 (m, 2H, CH2Cd), 1.80 (t, J ) 2.4 Hz, 3H, CH3),
1.57 (m, 2H, CH2), 1.02 (br, 1H, NH). 13C NMR (75 MHz, CDCl3):
δ 138.4, 114.6, 78.7, 77.2, 48.3, 38.6, 31.5, 29.0, 3.5. MS (relative
abundance): M+ (4), M+ - 1 (21), M+ - 2 (2), M+ + 1 (2), 122.1
(23), 108.0 (13), 94.0 (21), 82.0 (100), 70.0 (25), 53.0 (53), 42.0 (23).
HRMS: Calcd for C9H14N (M+ - H): 136.1126. Found: 136.1122.
Preparation of N-3-(Trimethylsilyl)-2-propynyl-4′-penten-1-
amine (21). A solution of 105.0 g (1.41 mol) of HCtCCH2Cl in 300
mL of THF was treated with n-BuLi (1.41 mol) at -78 °C with stirring
over a period of 2 h before 160.0 g (1.44 mol) of Me3SiCl was injected
into the reaction mixture. The solution was warmed to 25 °C, stirred
for 2 h, and then heated at 60 °C for 1 h. The reaction mixture was
cooled to 25 °C and poured into a separatory funnel containing 150
mL of H2O. After separation, the aqueous phase was extracted with 3
× 150 mL of CHCl3 and the combined extracts and organic phase were
dried over MgSO4, filtered, and concentrated by rotary evaporation.
Distillation (134-136 °C) afforded 128.5 g (62% yield) of Me3-
SiCtCCH2Cl as a colorless oil.
1H NMR (300 MHz, C6D6): δ 5.75-5.64 (m, 1H, CHd), 5.08-
4.89 (m, 2H, CH2d), 3.10 (d, 4J ) 2.4 Hz, 2H, NCH2Ct), 3.07 (dt, 4J
4
) 1.5 Hz, 2H, CH2N), 1.98 (t, J ) 2.3 Hz, 1H, HCt), 0.82 (br, 1H,
NH). 13C NMR (75 MHz, C6D6): δ 136.9, 115.9, 82.7, 71.5, 50.9,
37.4. MS (relative abundance): M+ (23), M+ + 1 (3), M+ - 1 (100),
80.0 (62), 68.0 (86), 56.1 (32), 52.0 (8), 39.0 (67). HRMS: Calcd for
C6H8N (M+ - H): 94.0657. Found: 94.0647.
Synthesis of N-n-Propylpropargylamine (27). A procedure similar
to that for 1 above was used to synthesize N-n-propylpropargylamine
by reaction of propargyl chloride (7.0 g, 0.092 mol) and n-propylamine
(36.0 g, 0.608 mol). Distillation (110-112 °C) of the residue gave
5.0 g (56% yield) of CH3CH2CH2NHCH2CtCH as a colorless oil.
1H NMR (300 MHz, C6D6): δ 3.11 (d, 4J ) 2.7 Hz, 2H, NCH2Ct),
2.40 (t, J ) 7.1 Hz, 2H, CH2N), 1.93 (t, 4J ) 2.4 Hz, 1H, HCt), 1.26
(m, 2H, CH2), 0.75 (t, J ) 7.5 Hz, 3H, CH3), 0.53 (br, 1H, NH). 13C
NMR (75 MHz, C6D6): δ 83.0, 71.1, 50.6, 38.3, 24.3, 11.9. MS
(relative abundance): M+ (12), M+ + 1 (1), M+ - 1 (2), 68.1 (100),
54.0 (7), 41.0 (29). HRMS: Calcd for C6H11N: 97.0891. Found:
97.0886.
Synthesis of N-Propargyl-4-penten-1-amine (29). A procedure
similar to that for 1 above was used to synthesize N-propargyl-4-penten-
1-amine from reaction of propargylamine (24.8 g, 0.449 mol) with
1-bromo-4-pentene (6.8 g, 0.0456 mol). Distillation (107-108 °C/55
Torr) of the residue gave 5.0 g (89% yield) of CH2dCHCH2CH2CH2-
NHCH2CtCH as a colorless oil.
1H NMR (300 MHz, CDCl3): δ 4.15 (s, 2H, CH2), 0.19 (s, 9H,
SiMe3).
A stirring solution of 12.2 g (0.14 mol) of 4-penten-1-amine in 70
mL of anhydrous THF was treated dropwise by syringe with 5.0 g
(0.034 mol) of ClCH2CtCSiMe3 at 0 °C over a period of 20 min. The
resulting mixture was stirred overnight at 25 °C before being poured
into a separatory funnel containing 200 mL of H2O. After separation,
the aqueous phase was extracted with 3 × 150 mL of diethyl ether.
The combined ethereal extracts and organic phase were washed with
brine (50 mL), dried over MgSO4, filtered, and concentrated by rotary
evaporation. Distillation (128-130 °C/30 Torr) gave 2.5 g (38% yield)
of the title compound as a colorless oil.
1H NMR (300 MHz, CDCl3): δ 5.81 (m, 1H, CHd), 4.96 (m, 2H,
CH2d), 3.41 (s, 2H, NCH2Ct), 2.66 (t, J ) 7.4 Hz, 2H, CH2N), 2.09
(m, 2H, CH2Cd), 1.57 (m, 2H, CH2), 1.19 (br, 1H, NH), 0.14 (s, 9H,
SiMe3). 13C NMR (100 MHz, CDCl3): δ 138.4, 114.7, 104.6, 87.7,
48.2, 39.2, 31.5, 29.0, 0.02. MS (relative abundance): M+ (12), M+
- 1 (12), M+ + 1 (4), 180.1 (41), 166.1 (10), 152.1 (16), 140.1 (100),
122.1 (40), 11.0 (48), 97.0 (14), 83.0 (49), 73.0 (41), 63.0 (3), 41.0
(16). HRMS: Calcd for C11H21NSi: 195.1443. Found: 195.1434.
Preparation of N-4-(Trimethylsilyl)-3-butynyl-4′-penten-1-amine
(23). A solution of 30.0 g (0.31 mol) of Me3SiCtCH in 200 mL of
THF was treated with n-BuLi (0.31 mol) at -78 °C with stirring over
a period of 2 h before 44.6 g (0.310 mol) of BrCH2CH2Cl was injected
into the reaction mixture. The solution was warmed to 25 °C and stirred
for 2 h, then heated at 60 °C for 2 h. The reaction mixture was next
cooled to 25 °C and poured into a separatory funnel containing 200
mL of H2O. After separation, the aqueous phase was extracted with 3
× 150 mL of CHCl3 and the combined extracts and organic phase were
dried over MgSO4, filtered, and concentrated by rotary evaporation.
Distillation (145-148 °C) afforded 7.6 g (16% yield) of Me3-
SiCtCCH2CH2Cl as a colorless oil.
1H NMR (300 MHz, C6D6): δ 5.72 (m, 1H, CHd), 4.96 (m, 2H,
CH2d), 3.12 (d, 4J ) 2.5 Hz, 2H, NCH2Ct), 2.46 (t, J ) 7.0 Hz, 2H,
CH2N), 1.93 (m, 2H, CH2Cd), 1.90 (m, 1H, HCt), 1.33 (m, 2H, CH2),
0.48 (br, 1H, NH). 13C NMR (75 MHz, C6D6): δ 138.8, 114.6, 82.9,
71.1, 48.1, 38.3, 31.7, 29.3. MS (relative abundance): M+ (1), M+
+
1 (1), M+ - 1 (9), 108.1 (3), 94.0 (8), 80.0 (17), 68.0 (100), 54.0 (5),
41.0 (30). HRMS: Calcd for C8H12N (M+ - H): 122.0970. Found:
122.0989.
Typical NMR-Scale Catalytic Hydroamination/Bicyclization Re-
actions. In the glovebox, the Cp′2SmCH(SiMe3)2 precatalyst (1.8 mg,
3.1 µmol) was loaded into an NMR tube equipped with a Teflon valve.
On the vacuum line, the NMR tube was evacuated and back-filled with
argon three times. Benzene (∼0.2 mL) was then vacuum transferred
into the tube. Next a mixture (∼0.4 mL) of benzene-d6 and N-allyl-
5-phenyl-4-pentyn-1-amine (1, 12.0 mg, 61.0 µmol) was added via
syringe under an argon flash while the tube which was maintained at
-78 °C to protect the precatalyst under the frozen benzene. The NMR
tube was next evacuated and back-filled with argon three times at -78
°C and finally sealed. The ensuing catalytic reaction was monitored
by 1H NMR. Upon completion, the reaction mixture was then
chromatographed on a small alumina column (200 mg, neutral) using
diethyl ether (7.0 mL) as eluant to remove the catalyst. The eluate
was concentrated by rotary evaporation to yield 8.1 mg (68% yield) of
1H NMR (300 MHz, CDCl3): δ 3.47 (t, J ) 7.4 Hz, 2H, CH2Cl),
2.56 (t, J ) 7.5 Hz, 2H, CH2), 0.05 (s, 9H, SiMe3).
A mixture of 20.0 g (0.23 mol) of 4-penten-1-amine and 2.4 g (0.015
mol) of Me3SiCtCCH2CH2Cl in a sealed storage tube was heated at
60 °C for 4 days before being poured into a separatory funnel containing
100 mL of H2O. After separation, the aqueous phase was extracted
with 3 × 50 mL of CH2Cl2. The combined extracts and organic phase
were washed with brine (50 mL), dried over MgSO4, filtered, and
concentrated by rotary evaporation. Distillation (138-140 °C/39 Torr)
gave 23 as a colorless oil (1.1 g, 35% yield).
1H NMR (300 MHz, C6D6): δ 5.69 (m, 1H, CHd), 4.95 (m, 2H,
CH2d), 2.53 (t, J ) 6.8 Hz, 2H, NCH2CH2Ct), 2.32 (t, J ) 7.1 Hz,
2H, CH2N), 2.20 (t, J ) 6.8 Hz, 2H, CH2Ct), 1.94 (m, 2H, CH2Cd),
1.34 (m, 2H, CH2), 0.63 (br, 1H, NH), 0.17 (s, 9H, SiMe3). 13C NMR
(75 MHz, C6D6): δ 138.9, 114.6, 106.3, 85.6, 48.8, 48.6, 31.8, 29.7,
21.6, 0.29. MS (relative abundance): M+ (2), 194.2 (3), 154.2 (7),
1
2 as a colorless oil. The products were identified by H, 13C, DEPT,
and 2D NMR as well as by GC/MS, and HRMS.
Typical Preparative-Scale Catalytic Hydroamination/Bicycliza-
tion Reactions. In the glovebox, 49.0 mg (85 µmol) of Cp′2NdCH-
(SiMe3)2 and 4.0 mL of C6H6 were loaded into a reaction vessel (30
mL) equipped with a magnetic stirbar. Next, 2.0 mL of a mixture of
N-allyl-4-hexyn-1-amine (3, 0.116 g, 847 µmol, 10-fold molar excess)
and benzene (∼1.9 mL) was added dropwise to the stirring precatalyst
solution. The resulting reaction mixture was then allowed to stir for 1
h at room temperature before the second 2.0 mL of the amine substrate

1762 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Li and Marks
mixture was added to the vessel. The procedure above was repeated
until 0.58 g of N-allyl-4-hexyn-1-amine (4235 µmol, 50-fold molar
excess) had been added to the vessel. The clear green solution was
then stirred at 25 °C for 3 more days. The reaction mixture was next
chromatographed on alumina (800 mg, neutral) using diethyl ether (30
mL) as the eluant to remove the catalyst. The eluate was concentrated
by rotary evaporation to yield 0.51 g (88% yield) of 1-methyl-2-
methylpyrrolizidine-8-ene (4) as a colorless oil. It was >95% pure by
1H, 13C NMR, and GC/MS.
The typical preparative-scale reaction procedure described above was
used in a reaction with Cp′2NdCH(SiMe3)2 (49.0 mg, 85 µmol) and 13
(0.280 g, 1879 µmol). The reaction mixture was chromatographed on
alumina (700 mg, neutral) using diethyl ether (20 mL) as the eluant to
remove the catalyst. The eluate was concentrated by rotary evaporation
to yield 0.26 g (92% yield) of 14 as a colorless oil.
Synthesis of cis- and trans-2-Methylpyrrolizidine (16a,b). These
pyrrolizidine derivatives18,20b,d were synthesized in both NMR- and
preparative-scale reactions. The Cp′2SmCH(SiMe3)2 precatalyst (5.7
mg, 9.8 µmol) in benzene-d6 (∼0.4 mL) and N-allyl-4-pentenamine
(58.8 mg, 470.0 µmïl, 47-fold molar excess) were used for the NMR-
scale reaction. The products were identified by 1H and 13C NMR and
by GC/MS. The NMR data agree well with the literature data.18,20c
13C NMR of cis-2-methylpyrrolizidine (16a) (75 MHz, 1H-decoupled,
C6D6): δ 64.6, 63.4, 56.7, 41.4, 34.0, 33.5, 27.3, 18.6. 13C NMR of
Synthesis of 1-Phenyl-2-methylpyrrolizidin-8-ene (2). This com-
pound was synthesized in both NMR and preparative-scale reactions.
The typical preparative-scale reaction procedure described above was
used with Cp′2NdCH(SiMe3)2 (49.0 mg, 85 µmol) and 1 (0.575 g, 2890
µmol). Flash chromatography (diethyl ether) yielded 2 as a colorless
oil (0.49 g, 85% yield).
1
trans-2-methylpyrrolizidine (16b) (75 MHz, H-decoupled, C6D6): δ
1H NMR (300 MHz, C6D6): δ 7.27-7.17 (m, 3H, Ph), 7.00-6.95
65.8, 64.5, 55.9, 42.9, 37.6, 32.9, 27.0, 18.0. The isomers exhibit
indistinguishable mass spectra. MS (relative abundance): M+ (55),
M+ - 1 (43), M+ + 1 (6), 110.0 (23), 97.0 (59), 83.0 (100), 68.0 (9),
55.0 (70), 41.0 (24) for trans-2-methylpyrrolizidine (ca. 55% yield by
1H NMR and GC). HRMS: Calcd for C8H15N: 125.1204. Found
125.1201. MS (relative abundance): M+ (45), M+ - 1 (30), M+ + 1
(64), 110.0 (19), 97.0 (59), 83.0 (100), 68.0 (9), 55.0 (81), 41.0 (31)
for cis-2-methylpyrrolizidine (ca. 45% yield by 1H NMR and GC).
HRMS: Calcd for C8H15N: 125.1204. Found: 125.1201.
3
2
(m, 2H, Ph), 3.38 (m, 1H, CH), 2.84 (dd, J ) 8.7 Hz, J ) 2.4 Hz,
3
2
1H, NCH2CH(ME)), 2.74 (dd, J ) 8.7 Hz, J ) 2.4 Hz, 1H, NCH2-
CHMe), 2.71 (m, 1H, NCH2), 2.20 (m, 1H, CH2Cd), 2.08 (m, 1H,
NCH2), 2.05 (m, 1H, CH2Cd), 1.84 (m, 1H, NCH2CH2), 1.73 (m, 1H,
NCH2CH2), 1.28 (d, J ) 6.90 Hz, 3H, CH3). 13C NMR (75 MHz,
C6D6): δ 153.5, 137.8, 129.1, 124.8, 123.5, 109.8, 60.0, 50.8, 43.3,
27.6, 25.1, 19.2. MS (relative abundance): M+ (47), M+ - 1 (9), M+
+ 1 (10), 184.1 (100), 156.1 (18), 128.1 (6), 115.0 (7). HRMS: Calcd
for C14H17N: 199.1361. Found: 199.1373.
The reaction was scaled up using the following procedure. In the
glovebox, 14.5 mg (25.0 µmol) of Cp′2SmCH(SiMe3)2 was loaded into
a reaction vessel (25 mL) equipped with a stirbar. Next, 2 mL of C6H6
was vacuum-transferred onto the precatalyst, followed by syringing 2.0
mL of a mixture of N-allyl-4-pentenamine (15, 0.21 g, 1680 µmol,
67-fold molar excess) and benzene (∼1.5 mL) onto the frozen benzene
and precatalyst mixture. The mixture was then freeze-pump-thaw
degassed and warmed to room temperature. The clear yellow solution
was stirred under argon for 5 days. Filtration followed by vacuum
transfer afforded a mixture of C6H6, cis-2-methylpyrrolizidine, and
trans-2-methylpyrrolizidine. After the benzene was removed by
distillation at atmospheric pressure, 194 mg (93% yield) of 16a,b (cis:
trans ratio ) 45:55 by comparison with lit.18,20c NMR data) was obtained
Synthesis of 1,2-Dimethylpyrrolizidin-8-ene (4). This pyrrolizidine
derivative was prepared in both NMR and preparative-scale reactions.
A procedure analogous to that for 2 above was used in the NMR-scale
reaction using Cp′2SmCH(SiMe3)2 (1.8 mg, 3.1 µmol) and N-allyl-4-
hexyn-1-amine (3, 11.0 mg, 80.7 µmol). The reaction mixture was
chromatographed on alumina (250 mg, neutral) using diethyl ether (5.0
mL) as eluant to remove the catalyst. The eluate was concentrated by
rotary evaporation to yield 8.2 mg (75% yield) of 2 as a colorless oil.
1H NMR (300 MHz, C6D6): δ 3.15 (t, J ) 8.0 Hz, 1H, N-CH2-
CH2), 2.99 (m, 1H, CH-Me), 2.62 (m, 1H, N-CH2), 2.33 (t, J ) 8.4
Hz, 1H, N-CH2-CH2), 2.27 (m, 1H, N-CH2), 1.99 (m, 2H, CH2-
Cd), 1.84 (m, 2H, N-CH2-CH2), 1.62 (s, 3H, CH3-Cd), 1.07 (d,
3H, CH3). 13C NMR (77 MHz, C6D6): δ 150.2, 104.3, 61.9, 52.9,
46.7, 27.3, 21.1, 18.6, 11.3. MS (relative abundance): M+ (36), M+
- 1 (23), M+ + 1 (4), 122.1 (100), 108.1 (7), 94.1 (16), 41.0 (7).
HRMS: Calcd for C9H15N: 137.1204. Found: 137.1241.
1
as a colorless oil. It was >95% pure by H NMR and GC/MS.
Synthesis of cis- and trans-2-Methylindolizidine (18a,b).
A
procedure similar to that used for 2 above in the NMR-scale reaction
was used for cis-2-methylindolizidine (18a) and trans-2-methyl-
indolizidine (18b). The Cp′2SmCH(SiMe3)2 precatalyst (3.0 mg, 5.2
µmol) in benzene-d6 (∼0.4 mL) and N-allyl-5-hexenamine (17, 8.8 mg,
64.0 µmïl) were used. The reaction mixture was chromatographed
on alumina (50 mg, neutral) using diethyl ether (7.0 mL) as eluant.
The eluate was concentrated by rotary evaporation to yield 7.0 mg (88%
yield) of 18 as a colorless oil.
Preparation of 1-Methyl-Z-2-ethylidenepyrrolizidin-8-ene (14)
and 1-Methyl-2-ethyl-6,7-dihydro-5H-pyrrolizine (14a). A procedure
analogous to that for 2 was used in the synthesis of bicyclic conjugated
diene 14 with Me2SiCp′′2NdCH(SiMe3)2 (2.9 mg, 4.8 µmol) and N-2-
butynyl-4′-hexyn-1-amine (13, 42.9 mg, 288.0 µmol). The yield (95%)
was estimated by 1H NMR and GC/MS after the product was isolated
from the catalyst by vacuum transfer. The title compound was
identified by NMR using1H, 13C, DEPT, 2D, and NOE difference
techniques.
1H NMR (300 MHz, C6D6): δ 4.80 (m, 1H, CHd), 3.60 (m, 2H,
NCH2Cd), 2.47 (t, J ) 6.6 Hz, 2H, NCH2), 2.00 (t, J ) 7.1 Hz, 2H,
CH2CN), 1.78 (m, 2H, CH2), 1.61 (t, 5J ) 1.1 Hz, 3H, CH3), 1.56 (td,
5J ) 2.1 Hz, 3H, CH3). 13C NMR (75 MHz, C6D6): δ 157.7, 150.1,
103.1, 100.9, 54.2, 50.1, 25.5, 21.3, 14.3, 9.1. MS (relative abun-
dance): M+ + 1 (7), M+ (44), M+ - 1 (18), 134.1 (100), 120.0 (8),
106.0 (8), 79.0 (7). HRMS: Calcd for C9H12N (M+ - CH3): 134.0970.
Found: 134.0945.
1
13C NMR of the major isomer 18a (85% yield by H NMR and
GC/MS; 75 MHz, C6D6): δ 63.9, 63.5, 53.3, 40.1, 31.6, 29.5, 26.1,
25.2, 20.8. The stereochemistry of trans-2-methylindolizidine was
assigned by comparison with literature chemical shift values.20a 13C
1
NMR of the minor isomer 18b (15% yield by H NMR and GC/MS;
75 MHz, C6D6): δ 65.4, 62.5, 53.3, 41.1, 31.8, 29.5, 26.1, 25.1, 23.2.
The isomers exhibit indistinguishable mass spectra. MS (relative
abundance) of the major isomer: M+ - 1 (100), M+ (47), M+ + 1 (4),
124.4 (16), 110.3 (34), 97.3 (78), 83.2 (16), 69.2 (24), 55.2 (16), 41.1
(25). High-resolution mass spectrum: Calcd for C9H16N (M+ - H):
138.1283. Found: 138.1278. MS (relative abundance) of the minor
isomer: M+ - 1 (100), M+ (43), 124.4 (15), 110.3 (32), 97.3 (74),
83.2 (14), 69.2 (20), 55.2 (16), 41.1 (23). HRMS: Calcd for C9H16N
(M+ - H): 138.1283. Found: 138.1278.
1-Methyl-(Z)-2-ethylidenepyrrolizidine-8-ene (14) was chromato-
graphed on silica gel using diethyl ether as the eluent to yield the
isomerization product 14a in 95% yield.
Synthesis of 2-Ethylidenepyrrolizidine (20). A 5-mm NMR tube
with a Teflon valve was charged with 2.8 mg (4.6 µmol) of Me2SiCp′′2-
NdCH(SiMe3)2 and N-2-butynyl-4′-penten-1-amine (19, 18.9 mg, 138.0
1H NMR (300 MHz, C6D6): δ 6.29 (s, 1H, CH), 3.27 (t, J ) 6.9
Hz, 2H, NCH2), 2.58 (q, J ) 7.5 Hz, 2H, CH2Me), 2.41 (t, J ) 7.1 Hz,
2H, CH2Cd), 2.11 (s, 3H, CH3), 1.80 (m, 2H, CH2), 1.30 (t, J ) 7.5
Hz, 3H, CH2CH3). 13C NMR (75 MHz, C6D6): δ 133.7, 129.5, 109.8,
107.1, 45.9, 27.6, 23.2, 19.9, 15.6, 10.0. MS (relative abundance):
M+ + 1 (7), M+ (14), 134.1 (100), 120.1 (6), 106.1 (7), 91.1 (6), 84.1
(11), 77.0 (6), 43.0 (12). HRMS: Calcd for C10H15N: 149.1205.
Found: 149.1210.
(20) (a) Stereochemistry assigned from NMR comparisons to similar
compounds: Heidt, P. C.; Bergmeier, S. C.; Pearson, W. H. Tetrahedron
Lett. 1990, 5441-5444. (b) Newcomb, M.; Deeb, T, M. J. Am. Chem. Soc.
1987, 109, 3163-3165. (c) Skvortsov, I. M.; Antipova, I. V. J. Org. Chem.
USSR 1979, 15, 777-783. (d) Surzur, J.-M.; Stella, L. Tetrahedron Lett.
1974, 2191-2194.

Catalytic Tandem C-N and C-C Bond-Forming Reactions
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1763
µmol) in 0.6 mL of C6D6 as described above, and the mixture was
maintained at 25 °C. The progress of the reaction was monitored by
1H NMR. The products were vacuum transferred via a short Y-type
connecting tube to give a mixture of 20, CH2(SiMe3)2, and C6D6. The
title compound was identified by NMR spectroscopy including 1H, 13C,
DEPT, 2D, and NOE difference spectroscopy. The yield of 20 (92%)
reaction was monitored by 1H NMR. The yield of 28 (95%) was
estimated by 1H NMR and GC/MS after vacuum transfer of the
volatiles.
1H NMR (300 MHz, C6D6): δ 6.29 (s, 1H, CHd), 6.02 (s, 1H,
CHd), 3.63 (t, J ) 7.2 Hz, 2H, CH2N), 3.54 (d, J ) 5.7 Hz, 2H,
C-CH2-NH), 2.39 (m, 2H, CH3CH2CH2NH), 2.24 (s, 3H, CH3), 1.54
(m, 2H, CH2CH2N), 1.30 (m, 2H, CH2CH2NH), 0.80 (t, J ) 7.1 Hz,
CH3CH2CH2NH), 0.72 (t, J ) 7.5 Hz, CH3CH2CH2N), 0.50 (br, 1H,
NH). 13C NMR (75 MHz, C6D6): δ 129.3, 119.1, 116.9, 109.7, 51.5,
48.0, 46.4, 25.1, 23.5, 12.4, 11.9, 11.4. MS (relative abundance): M+
(27), M+ + 1 (5), M+ - 1 (13), 136.1 (100), 120.1 (11), 108.1 (6),
94.1 (28), 41.0 (7). HRMS: Calcd for C12H22N2: 194.1783. Found:
194.1779.
1
was estimated by H NMR and GC/MS.
1H NMR (300 MHz, C6D6): δ 5.23 (m, 1H, CHd), 3.67 (d, J )
15.0 Hz, 1H), 3.38 (m, 1H), 3.05 (d, J ) 14.7 Hz, 1H), 2.97 (m, 1H),
2.45 (m, 2H), 1.90 (m, 1H), 1.61 (m, 4H), 1.45 (d, J ) 6.9 Hz, 3H,
CH3). 13C NMR (75 MHz, C6D6): δ 142.4, 114.7, 64.6, 55.7, 54.8,
38.7, 31.2, 25.7, 14.7. MS (relative abundance): M+ - 1 (69), M+
(100), M+ + 1 (11), 122.1 (73), 109.1 (35), 94.1 (25), 83.1 (22), 67.1
(39), 55.1 (31), 49.0 (20), 41.0 (46), 36.0 (20). HRMS: Calcd for
C9H15N: 137.1204. Found: 137.1203.
Synthesis of 4-Methyl-2-((5-pentenylamino)methyl)-1-(4-pen-
tenyl)pyrrole (30). A 5-mm NMR tube with a Teflon valve was
charged with 2.0 mg (3.5 µmol) of Cp′2SmCH(SiMe3)2 and N-propargyl-
4-penten-1-amine (5, 3.48 mg, 28.3 µmol) in 0.6 mL of C6D6 as
described above, and the reaction mixture was heated at 60 °C. The
Synthesis of 2-((Trimethylsilyl)methylidene)pyrrolizidine (22). A
5-mm NMR tube with a Teflon valve was charged with 3.3 mg (5.5
µmol) of Me2SiCp′′2NdCH(SiMe3)2 and N-3-(trimethylsilyl)-2-propyn-
4′-penten-1-amine (21, 19.3 mg, 99.0 µmol) in 0.6 mL of C6D6 as
described above, and the reaction mixture was allowed to stand at 25
1
progress of the reaction was monitored by H NMR. The yield of 30
1
(95%) was estimated by H NMR and GC/MS after vacuum transfer
of the volatile products.
1
°C. The progress of the reaction was monitored by H NMR. The
1H NMR (300 MHz, C6D6): δ 6.29 (s, 1H, CHd), 6.01 (s, 1H,
CHd), 5.68 (m, 2H, 2CHd), 4.97 (m, 4H, 2CH2d), 3.67 (t, J ) 7.2
Hz, 2H, CH2N), 3.51 (d, J ) 6.3 Hz, 2H, C-CH2-NH), 2.42 (m, 2H,
CH2NH), 2.23 (s, 3H, CH3), 1.93 (m, 2H, CH2), 1.86 (m, 2H, CH2),
1.61 (m, 2H, CH2), 1.37 (m, 2H, CH2), 0.92 (br, 1H, NH). 13C NMR
(75 MHz, C6D6): δ 138.8, 138.0, 130.8, 119.0, 117.0, 114.9, 114.5,
109.8, 48.9, 46.3, 45.7, 31.8, 31.0, 30.9, 29.6, 12.3. MS (relative
abundance): M+ (3), M+ - 1 (10), 206.8 (6), 161.9 (100), 133.9 (16),
108.0 (29), 94.0 (16), 41.0 (8). HRMS: Calcd for C16H26N2: 246.2096.
Found: 246.2103.
title compound was identified by NMR spectroscopic analysis including
1H, 13C, DEPT, 2D, and NOE difference spectroscopy. The yield of
1
22 (80%) was estimated by H NMR and GC/MS after the product
was isolated by vacuum transfer of the volatiles.
1H NMR (300 MHz, C6D6): δ 5.48 (m, 1H, CH)), 3.77 (d, J )
15.0 Hz, 1H), 3.40 (m, 1H), 3.19 (d, J ) 15.0 Hz, 1H), 2.97 (m, 1H),
2.47 (m, 2H), 2.14 (m, 1H), 1.63 (m, 4H), 0.46 (s, 9H, SiMe3). 13C
NMR (100 MHz, C6D6): δ 147.1, 119.2, 63.9, 58.6, 54.7, 42.7, 31.4,
25.7, -0.20. MS (relative abundance): M+ - 1 (22), M+ (20), M+
+
1 (4), 180.1 (19), 166.0 (14), 152.1 (11), 122.1 (100), 111.0 (7), 94.1
(11), 83.0 (23), 73.0 (66), 59.0 (14), 45.0 (8). HRMS: Calcd for C11H21-
NSi: 195.1443. Found: 195.1436.
Synthesis of cis-1H,2H,3H,5H,10H-2,7-Dimethyldipyrrolo[1,2-a:
1′,2′-d]pyrazine (31). A 5-mm NMR tube with a Teflon valve was
charged with 3.0 mg (5.2 µmol) of Cp′2SmCH(SiMe3)2 and N-
allylpropargylamine (1, 3.4 mg, 36 µmol) in 0.6 mL of C6D6 as
described above, and the mixture was heated at 60 °C. The progress
of the catalytic reaction was monitored by 1H NMR. The title
compound was identified by GC/MS and NMR spectroscopic analysis
including 1H, 13C, DEPT, 2D, and NOE difference spectroscopy. The
reaction mixture was chromatographed on alumina (50 mg, neutral)
using diethyl ether (5.0 mL) as the eluant to remove the catalyst. The
eluate was concentrated by rotary evaporation to yield 3.2 mg (93%
yield) of the title compound as a colorless oil.
Synthesis of 2-Methyl-1-(4-(trimethylsilyl)-3-butynyl)pyrrolidine
(24). This cyclized pyrrolidine derivative was prepared using 3.3 mg
(5.5 µmol) of Me2SiCp′′2NdCH(SiMe3)2 and N-4-(trimethylsilyl)-3-
butynyl-4′-penten-1-amine (23, 18.4 mg, 88.0 µmol) in 0.6 mL of C6D6
1
as described above. The yield (>90%) was estimated by H NMR
and GC/MS after vacuum transfer of the volatiles.
1H NMR (400 MHz, C6D6): δ 2.94 (m, 2H), 2.38 (t, J ) 7.5 Hz,
2H), 2.26 (m, 1H), 2.09 (m, 1H), 1.86 (m, 1H), 1.59 (m, 2H), 1.36 (m,
1H), 1.21 (m, 1H), 0.94 (d, J ) 6.0 Hz, 3H, Me), 0.22 (s, 9H, SiMe3).
13C NMR (100 MHz, C6D6): δ 106.4, 84.7, 59.1, 53.4, 52.7, 32.8,
21.8, 20.3, 19.0, 0.08. MS (relative abundance): M+ (1), M+ - Me
(11), 98.0 (100), 69.0 (12), 41.0 (9). HRMS: Calcd for C11H20NSi
(M+ - Me): 194.1365. Found: 194.1374.
1H NMR (300 MHz, C6D6): δ 6.20 (s, 1H, CHd), 5.88 (s, 1H,
CHd), 4.00 (d, J ) 13.2 Hz, 1H), 3.43 (d, J ) 3.9 Hz, 1H), 3.39 (d,
J ) 3.9 Hz, 1H), 3.30 (t, J ) 10.7 Hz, 1H), 3.14 (d, J ) 13.2 Hz, 1H),
3.06 (d, J ) 8.0 Hz, 1H), 2.26 (s, 3H, CH3), 2.19-2.13 (m, 2H), 1.56
(t, J ) 8.9 Hz, 1H), 1.09 (m, 1H), 0.79 (d, J ) 6.9 Hz, 3H, CH3). 13
C
Synthesis of 4-Methyl-2-((2-allylamino)methyl)-1-allylpyrrole
(26). A 5-mm NMR tube with a Teflon valve was charged with 3.0
mg (5.2 µmol) of Cp′2SmCH(SiMe3)2 and N-allylpropargylamine (25,
35.4 mg, 372.2 µmol) in 0.6 mL of C6D6 as described above, and the
mixture was heated at 60 °C. The progress of the reaction was
NMR (75 MHz, C6D6): δ 129.3, 118.0, 116.6, 104.6, 63.2, 60.3, 51.0,
50.5, 36.7, 30.8, 20.1, 12.5. MS (relative abundance): M+ (64), M+
+ 1 (10), M+ - 1 (69), 175.1 (6), 133.1 (9), 107.1 (100), 94.1 (7),
79.0 (6), 41.0 (7). HRMS: Calcd for C12H18N2: 190.1470. Found:
190.1450.
1
monitored by H NMR. The title compound was identified by GC/
Synthesis of N-Allyl((1′-octen-3′-yn-2′-yl)methyl)methanamine
(32). A 5-mm NMR tube with a Teflon valve was charged with 3.0
mg (5.2 µmol) of Cp′2SmCH(SiMe3)2, N-allylpropargylamine (1, 5.4
mg, 56.9 µmol), and 1-hexyne (4.25 mg, 52.0 µmol) in 0.6 mL of C6D6
as described above, and the mixture was heated at 60 °C. The progress
MS and NMR including 1H, 13C, DEPT, 2D, and NOE difference
1
spectroscopy. The yield of 26 (92%) was estimated by H NMR and
GC/MS after vacuum transfer of the volatile products.
1H NMR (300 MHz, C6D6): δ 6.29 (s, 1H, CHd), 6.01 (s, 1H,
CHd), 5.85-5.66 (m, 2H, 2CHd), 5.12-4.78 (m, 4H, 2CH2d), 4.26
(m, 2H, N-CH2-CHd), 3.49 (d, J ) 6.3 Hz, 2H, C-CH2-NH), 3.03
(m, 2H, NH-CH2-CHd), 2.21 (s, 3H, CH3), 0.68 (br. 1H, NH). 13C
NMR (75 MHz, C6D6): δ 136.1, 130.7, 128.5, 119.6, 117.2, 115.3,
115.2, 110.3, 51.8, 48.9, 45.4, 12.3. MS (relative abundance): M+
(33), M+ + 1 (5), M+ - 1 (17), 148.1 (10), 134.1 (100), 120.1 (19),
107.1 (18), 94.1 (18), 80.0 (6), 68.1 (12), 41.0 (30). HRMS: Calcd
for C12H18N2: 190.1470. Found: 190.1458.
1
of the reaction was monitored by H NMR. The title compound was
1
identified by GC/MS and NMR spectroscopic studies including H,
13C, DEPT, and NOE difference spectroscopy. The yield of 32 (39%)
1
was estimated by H NMR and GC/MS after vacuum transfer of the
volatiles. This reaction also generates 31 (22%) and 2-butyl-1-octen-
3-yne4a (39%).
1H NMR (300 MHz, C6D6): δ 5.89-5.79 (m, 1H, CHdCH2), 5.47
(d, J ) 1.8 Hz, 1H, CH2d), 5.29 (d, J ) 1.8 Hz, 1H, CH2d), 5.18-
4.97 (m, 2H, CH2d), 3.25 (s, 2H, C-CH2-NH), 3.09 (d, J ) 6.0 Hz,
2H, HN-CH2-CHd), 2.11 (t, J ) 6.9 Hz, 2H, CH2-C), 1.29 (m,
4H, 2CH2), 0.76 (t, J ) 7.2 Hz, 3H, CH3), 0.45 (br, 1H, NH). 13C
NMR (75 MHz, C6D6): δ 137.3, 131.7, 119.3, 115.0, 91.1, 81.7, 54.1,
Synthesis of 4-Methyl-2-((3-propylamino)methyl)-1-propyl-
pyrrole (28). A 5-mm NMR tube with a Teflon valve was charged
with 3.0 mg (5.2 µmol) of Cp′2SmCH(SiMe3)2 and N-n-propylpropar-
gylamine (3, 64.9 mg, 669.1 µmol) in 0.6 mL of C6D6 as described
above, and the mixture was heated at 60 °C. The progress of the
51.0, 30.8, 21.9, 18.9, 13.4. MS (relative abundance): M+ (5), M+
+

1764 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Li and Marks
Table 1. Intramolecular Hydroamination/Bicyclization Results
Scheme 2. Proposed Pathway for Organolanthanide-
Catalyzed Sequential C-N and C-C Bond Formation
1 (1), M+ - 1 (3), 150.2 (3), 134.1 (20), 120.1 (33), 70.1 (100), 41.1
(34). HRMS: Calcd for C12H19N: 177.1517. Found: 177.1521.
Kinetic Study of Hydroamination/Bicyclization. In a typical
experiment, an NMR sample was prepared as described in the typical
NMR-scale catalytic reaction section but the tube was maintained at
-78 °C until kinetic measurements were initiated. Before the measure-
ments, the spectrometer probe was equilibrated at the appropriate
temperature (T ( 0.2 °C, checked with a methanol or ethylene glycol
temperature standard). The NMR tube was then quickly warmed with
shaking for ∼8 s and inserted into the probe. Data were acquired using
4-8 scans per time interval with a long pulse delay to avoid saturation
of the signal. The kinetics were usually monitored by the intensity
changes in the substrate resonances over three or more half-lives. The
substrate concentration (C) was measured from the area (AS) of integrals
a Cp′2SmCH(SiMe3)2 as precatalyst. b Me2SiCp′′2NdCH(SiMe3)2 as
precatalyst. c NMR-scale reaction with the yield determined by 1H
NMR and GC/MS after vacuum transfer of volatile products. d NMR-
scale reaction and isolated yield. e Preparative-scale reaction and
isolated yield. f Data from ref 7a.
tion of the scope, stereoselectivity, lanthanide ion sensitivity,
ancillary ligand sensitivity, kinetics, and mechanism of these
organolanthanide-catalyzed tandem C-N, C-C bond-forming
transformations. The intramolecular hydroamination/bicycliza-
tions of aminodialkynes, aminodialkenes, and aminoalkenynes
are discussed first, followed by the intermolecular hydro-
amination/cyclizations of aminoalkynes. This study capitalized
upon our previous success in organolanthanide-catalyzed carbon-
heteroatom bond formation6,7 and upon well-characterized
organolanthanide-catalyzed carbon-carbon bond formation
processes.3,4
1
of the H-normalized CH2d signal, standardized to the area (Al) of
free CH2(SiMe3)2, which is quantitatively generated by reaction of the
precatalysts with the substrate (Scheme 2). All data could be
convincingly fit (R > 0.98) by least-squares to eq 3, where CO (CO
)
A
SO/Al0) is the initial concentration of substrate (relative to precatalyst)
and C (AS/A1) is the substrate concentration at time, t. The ratio of
catalyst to substrate was
Scope of Organolanthanide-Catalyzed Intramolecular
Bicyclization. The organolanthanides Cp′2LnCH(SiMe3)2 and
Me2SiCp′′2LnCH(SiMe3)2 are precatalysts for intramolecular
hydroamination/bicyclization of aliphatic and aromatic amino-
dialkynes, aminodialkenes, and aminoalkenynes to yield the
corresponding monocyclic and/or bicyclic organonitrogen com-
pounds as shown in Table 1, where Nt is the catalytic turnover
frequency at the temperature indicated. The catalytic bicycliza-
tion reactions in general proceed to completion at room
temperature under inert atmosphere and are conveniently
monitored by 1H NMR spectroscopy. As in the case of
organolanthanide-catalyzed hydroamination/cyclization of
aminoolefins6b,c and aminoalkynes,7a the reaction of Cp′2NdCH-
(SiMe3)2,, Cp′2SmCH(SiMe3)2, and Me2SiCp′′2SmCH(SiMe3)2
with the present substrates results in distinct color changes
concurrent with catalytic initiation and termination. The original
green and orange solutions of the neodymium and samarium
alkyl precatalysts, respectively, in C6D6 or C7D8 instantaneously
turn to the characteristic blue and yellow colors of the
mt ) (CO - C)
(3)
accurately measured from the ratio of ASO and Al0 or from the area of
product and A1. The absolute concentrations of catalyst and substrate
were determined with an internal FeCp2 standard. The turnover
frequency (Nt, h-1) was calculated from the least-squares determined
slope (m) of the resulting plot. Typical initial substrate concentrations
were in the range 0.13-0.60 M, and typical catalyst concentrations in
the range 3.0-17 mM.
Results
The principal goal of this study was to exploit the possibility
of coupling sequences of organolanthanide-mediated element-
element bonding-forming reactions into catalytic cycles for
construction of polycyclic heteroatom-containing frameworks
(e.g., pyrrolizidines, indolizidines, pyrroles, pyrazines, and other
frameworks).8-10 A closely related objective involved examina-

Catalytic Tandem C-N and C-C Bond-Forming Reactions
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1765
corresponding amine-amido complexes6b,c with initiation of
catalytic turnover. Upon consumption of the amine substrates,
the resulting reaction solutions return to the original colors. The
pyrrolidine, pyrrolizidine, and indolizidine products were char-
acterized by NMR spectroscopy (1H, 13C, and 2D), GC/MS,
and high-resolution mass spectroscopy or by comparison with
literature 1H/13C NMR spectral data and data for authentic
samples. Isolation procedures for products involved chroma-
tography, or high vacuum transfer of the products and other
volatiles followed by removal of solvent and CH2(SiMe3)2.
Preparative-scale reactions were carried out in sealed tubes,
affording isolated products in 52-93% yields as shown in Table
1. All products in NMR or preparative-scale reactions were
separated from the catalysts and were >95% pure by 1H NMR
and GC/MS.
The present catalytic process, as illustrated in eq 4, for the
transformations of unsaturated secondary amines, may in
principle generate bicyclic or undesired monocyclic heterocycles
(due to premature Ln-C protonolysis). However, the results
of the cyclizations summarized in Table 1 demonstrate that
C-N/C-C fusions can be regiospecifically coupled in sequence
to exclusively assemble bicyclic rather than monocyclic products
indicate that both internal alkynes (entries 1, 2, and 7) and
terminal alkenes (entries 8-11) undergo addition to Ln-N
bonds, followed by insertion of a second olefin (entries 1, 2, 8,
and 9) or alkyne (entries 7, 10, and 11) into the resulting Ln-C
bonds to yield bicyclized pyrrolizidine (entries 1, 2, 7, 8, 10,
and 11) and indolizidine (entry 9) skeletons. Entries 7, 10, and
11 illustrate that internal alkynes are effective insertive groups
yielding exo-alkene functionalized pyrrolizidines. Entry 7
indicates rapid sequential alkyne/alkyne insertive bicyclization
to introduce two regions of heterocyclic unsaturation. Entries
8 and 9 illustrate alkene/alkene bicyclization to produce known
saturated pyrrolizidine 1618,20b-d and indolizidine 18 (cis:trans
ratios ) 45:55 and 85:15,20a respectively).
Attempts to couple terminal alkyne/alkene insertions in
substrates 5 and 11 (entries 3 and 6) to form bicyclized
compounds were unsuccessful, and monocyclized pyrrolidines,
the products of competing, premature protonolysis (eq 4), are
obtained instead. It is likely that closure to form a seven-
membered ring is kinetically disfavored in 11, judging from
prior aminoalkene results. Entries 4 and 5 also reveal that a
second (olefinic) insertion into an R-(trimethylsilyl)vinyl-
lanthanide bond is also impeded; instead a slow, precedented
catalytic3k double-bond migration occurs in the case of entry 4.
Entry 12 illustrates that the second ring-closing reaction fails
for alkyne insertion into an Ln-C bond to form a six-membered
ring, and a monocyclic structure is instead preferentially
generated.
In regard to reaction rate, the present disubstituted amine
bicyclizations are comparable in turnover frequency to the
corresponding cyclizations of primary aminoalkenes and amino-
alkynes for the same catalysts, temperatures, and reaction
conditions.6,7 For example, entry 2 illustrates that N-allyl-4-
hexyn-1-amine is bicyclized regiospecifically to the correspond-
ing pyrrolizidine derivative with Nt as high as 777 h-1 at room
temperature. In comparison, the turnover frequencies for the
cyclization of 4-pentyn-1-amine and 4-penten-1-amine to the
corresponding heterocycles are 580 h-1 and 6 h-1, respectively
(eqs 5 and 6).6b,c,7a It can be seen in Table 1 that substrates
lacking alkynyl functionalities undergo bicyclization with more
modest Nt values (entries 8 and 9). In addition, in two directly
when catalysts with less coordinative unsaturation at the metal
center (e.g., Cp′2LnCH(SiMe3)2-type complexes) are used at
room temperature (Table 1, entries 1, 2, 7, 8, 9, 10, and 11).
However, small amounts (3-15%) of monocyclic byproducts
are observed when more “open” Me2SiCp′′2LnCH(SiMe3)2
precatalysts and/or high substrate concentrations are employed
and at slightly elevated temperatures (e.g., 60 °C). In lieu of
this observation, optimization of the catalytic reaction conditions
was carried out using the 3 f 4 transformation. It is found
that using less “open” organolanthanide catalysts, relatively low
operating substrate:catalyst molar ratios (substrate:catalyst )
12-60), and carrying out the reactions at room temperature
ensures regiospecific pendant olefin insertion to close the
second ring in the catalytic cycle (e.g., entry 2, in Table 1;
Scheme 2). Similar effects are found in the insertion of the
terminal olefin into the R-phenylvinyl-lanthanide bond in the
bicyclization of N-allyl-5-phenyl-4-pentyn-1-amine (entry 1).
It can be seen that the present process, using the aforemen-
tioned optimized conditions, effects efficient coupling of alkyne/
alkene (entries 1 and 2), alkene/alkene (entries 8 and 9), alkene/
alkyne (entries 10 and 11), and alkyne/alkyne (entry 7) moieties
in catalytic cycles, by the insertion of carbon-carbon multiple
bonds into the Ln-N and Ln-C bonds, in sequence. This
catalytic reaction is capable of forming five-five and six-five
polycyclic skeletons (entries 1, 2, and 7-11). The results
competitive intramolecular cyclizations, a distinct preference
for alkyne insertion over alkene insertion is observed (9 f 10,
11 f 12; eqs 7 and 8). This is in accord with earlier
regiospecificity observations on organolanthanide-mediated ole-
fin and alkyne hydroamination/cyclization and is not surprising
in view of the known activating effects of SiMe3 substitution,21
the greater exothermicity of additions to alkynes versus those
to alkenes,22-24and the enhanced nucleophilic reactivity of
alkynes over alkenes.25
Substrate Substituent, Lanthanide, and Ancillary Ligand
Effects on the Bicyclization Process. Kinetic and mechanistic
studies of the Cp′2Ln-catalyzed hydroamination/bicyclization
reactions to be discussed below argue that the turnover-limiting
(21) (a) Bassindale, A. R.; Taylor, P. G. In The Chemistry of Organic
Silicon Compounds; Patai, S., Rappaport, Z., Eds.; Wiley: Chichester, U.K.,
1989; Chapter 14. (b) Fleming, I. In ComprehensiVe Organic Chemistry;
Jones, N. D., Ed.; Pergamon Press: Oxford, U.K., 1979; Chapter 13.

1766 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Li and Marks
step is the intramolecular insertion of an unsaturated carbon-
carbon bond into the Ln-N bond, followed by a more rapid
second ring closure and/or protonolysis of the intermediate
Ln-C bond (Scheme 2, eq 4). In analogy to earlier observations
on the hydroamination/cyclization of aminoolefins and amino-
alkynes mediated by organolanthanides,6,7 the transition states
involved in the first ring-closing step for multifunctional
substrates should be both electronically and sterically sensitive
to the environment of the four-membered insertive transition
state configuration.26 Entries 1, 2, and 7 of Table 1 indicate
that the first ring-closing process involving CtC addition to
the Ln-N bond is considerably more rapid than that for alkenes
(entries 8-11). These results are consistent with previous
observations on organolanthanide-catalyzed cyclization of ami-
noalkenes and aminoalkynes.6,7 For diolefinic amines, that
bicyclization of N-allyl-4-penten-1-amine (entry 8) is more rapid
under the same conditions than that of N-allyl-5-hexen-1-amine
(entry 9) is consistent with initial insertion via a preferred quasi-
seven-membered transition state in the former (e.g., A). Similar
ring size-reaction rate relationships prevail in organolanthanide-
mediated aminoolefin cyclizations.6c
Figure 1. Plot of turnover frequency (Nt) as a function of eight-
coordinate Ln3+ ionic radius for the intramolecular hydroamination/
bicyclization of PhCtCCH2CH2CH2NHCH2CHdCH2 (1 f 2) in
benzene-d6 using Cp′2LnCH(SiMe3)2 complexes as the precatalysts.
Table 2. Lanthanide Ion Size Effects on the Turnover Frequency
for Hydroamination/Bicyclization
a Eight-coordinate ionic radii from ref 27.
In regard to substrate substituents, Table 1 reveals that the
rate of bicyclization is decreased by sterically encumbered
substituents for the same alkyne/alkene (entries 1 and 2) and
alkene/alkyne (entries 8 and 10) frameworks. It can also be
seen that the course of bicyclization depends on the pattern of
amine substitution. Attempts to effect six- or seven-membered
ring closure with initial cyclization via insertion of pendant γ
or δ unsaturation were unsuccessful (entries 5, 6, and 12),
mirroring trends noted previously for monofunctional sub-
strates.6,7
(22) Requisite experimental thermochemical data are not available to
compare the thermodynamics of amine addition to olefins versus that to
alkynes. However, ∆H for NH3 addition to ethylene can be estimated from
thermochemical data23,24 to be -13 kcal/mol. Calculations at the AM-1
level place NH3 addition to acetylene (to yield CH3CHdNH) as 17 kcal/
mol more exothermic than to ethylene. Furthermore, the addition of CH4
to acetylene is estimated from thermochemical data23,24 to be ∼14 kcal/
mol more exothermic and ∼14 kcal/mol more exergonic than that to
ethylene.
In regard to metal ion size effects, interestingly, and in marked
contrast to the analogous hydroamination/cyclization of amino-
alkynes,7 the present results illustrate an acceleration rather than
a deceleration in rate when lanthanides with larger ionic radii27
are used as the catalysts. Thus, the turnover frequency for
conversion of N-allyl-5-phenyl-4-pentyn-1-amine to 1-phenyl-
2-methylpyrrolizidin-8-ene (1 f 2) under identical reaction
conditions increases dramatically from the smallest Lu3+ (0.977
Å), to intermediate Sm3+ (1.079 Å), to Nd3+ (1.109 Å), and to
the largest La3+ (1.160 Å) (Figure 1; Table 2). Similar trends
have been observed in a variety of Ln-catalyzed processes where
the turnover-limiting step is olefin insertion.3c,d,6b,c In the case
of organolanthanide-catalyzed aminoolefin hydroamination/
cyclization, opening the metal coordination sphere by connecting
ancillary Cp ligands (Cp′2Ln f Me2SiCp′′2Ln or Me2SiCp′′-
(R*Cp)Ln)6b,12b results in higher turnover frequencies, presum-
ably reflecting steric demands in turnover-limiting olefin
insertion step (e.g., A). However, in the present bicyclizations,
a decrease in Nt for aminodiolefin substrates is observed for
Me2SiCp′′2NdCH(SiMe3) as the precatalyst (entry 8). A similar
(23) Metal-ligand bond enthalpies from: (a) Nolan, S. P.; Stern, D.;
Hedden, D.; Marks, T. J. ACS Symp. Ser. 1990, 428, 159-174. (b) Nolan,
S. P.; Stern, D.; Marks, T. J. J. Am. Chem. Soc. 1989, 111, 7844-7853. (c)
Schock, L. E.; Marks, T. J. J. Am. Chem. Soc. 1988, 110, 7701-7715. (d)
Bruno, J. W.; Marks, T. J.; Morss, L. R. J. Am. Chem. Soc. 1983, 105,
6824-6832. (e) Reference 6d.
(24) Organic fragment bond ethalpies from: (a) Griller, D.; Kanabus-
Kaminska, J. M.; Maccoll, A. J. Mol. Struct. 1988, 163, 125-131. (b)
McMillan, D. F.; Golden, D. M. Annu. ReV. Phys. Chem. 1982, 33, 493-
532 and references therein. (c) Benson, S. W. Thermochemical Kinetics,
2nd ed.; John Wiley and Sons: New York, 1976; Appendix Tables A.10,
A.11, A.22. (d) Benson, S. W. J. Chem. Educ. 1965, 42, 502-518. (e)
Pedley, J. B.; Naylor, R. D.; Kirby, S. P. Thermochemical Data of Organic
Compounds, 2nd ed.; Chapman and Hall: London, 1986; Appendix Tables
1 and 3.
(25) For a comparison of nucleophilic addition chemistry involving
alkenes and alkynes, see for example: Lowry, T. H.; Richardson, K. S.
Mechanism and Theory in Organic Chemistry, 3rd ed.; Harper & Row:
New York, 1987; Chapter 7 and references therein.
(26) (a) Eshuis, J. J. W.; Tan, Y. Y.; Meetsma, A.; Teuben, J. H.;
Renkema, J.; Evens, G. G. Organometallics 1992, 11, 362-369. (b) Piers,
W. E.; Shapiro, P. J.; Bunel, E. E.; Bercaw, J. E. Synlett 1990, 74-84. (c)
Burger, B. J.; Thompson, M. E.; Cotter, W. D.; Bercaw, J. E. J. Am. Chem.
Soc. 1990, 112, 1566-1577. (d) Thompson, M. E.; Baxer, S. M.; Bulls, A.
R.; Burger, B. J.; Nolan, M. C.; Santarsiero, B. D.; Schaefer, W. P.; Bercaw,
J. E. J. Am. Chem. Soc. 1987, 109, 203-219. (e) Doherty, N. M.; Bercaw,
J. E. J. Am. Chem. Soc. 1985, 107, 2670-2682.
(27) Shannon, R. D. Acta Crystallogr. 1976, A32, 751-760.

Catalytic Tandem C-N and C-C Bond-Forming Reactions
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1767
Scheme 3. Proposed Pathway for Organolanthanide-
Catalyzed Intermolecular Hydroamination/Cyclization
Figure 2. Plot of CH2dCHCH2CH2CH2NHCH2CHdCH2 concentra-
tion as a function of time at constant catalyst concentration for the
hydroamination/ bicyclization of CH2dCHCH2CH2CH2NHCH2CHdCH2
(15 f 16) using Cp′2SmCH(SiMe3)2 as the precatalyst in benzene-d6.
The line represents least-squares fit to the data points.
constant (k) for the 15 f 16 conversion at 21 °C is 0.011(2)
s-1
.
Figure 3. Plot of reaction rate vs catalyst concentration for the
intramolecular hydroamination/bicyclization of CH2dCHCH2CH2CH2-
NHCH2CHdCH2 (15 f 16) using Cp′2SmCH(SiMe3)2 as the pre-
catalyst in benzene-d6. The line represents a least-squares fit to the
data points.
υ ) k[substrate]0[Sm]1
(9)
In the presence of the Nd3+ (4f3) and Sm3+ (4f5) catalysts,
both paramagnetically broadened 1H NMR substrate and product
resonances are observed during turnover. This behavior is
analogous to that of aminoolefin and aminoalkyne cyclizations
where amine-amido Cp′2LnNHR(NH2R)x complexes undergo
bound amine-amide and bound-free amine (substrate and
product) exchange, which is fast on the NMR time scale at 25
°C.6b,c
relationship between Nt and ancillary ligation-based Ln3+
unsaturation is observed in bicyclization of internal alkynes and
terminal olefins (entries 1 and 2). However, as in the case of
aminoolefin cyclization,6b,c increasing Nt values with more open
coordination spheres are observed in bicyclization of apparently
more sterically demanding internal alkyne/alkyne and terminal
olefin/internal alkyne substrates (entries 7 and 10).
Organolanthanide-Catalyzed Intermolecular Hydroamin-
ation/Cyclization. Coupled intermolecular hydroamination and
intramolecular cyclizations, mediated by organolanthanides, are
also efficiently incorporated into catalytic cycles. The efficiency
of such processes was first investigated with substituted
secondary propargylamines (eq 10). In a typical intermolecular
hydroamination/cyclization procedure (see the Experimental
Section for details), a 70:1 molar ratio of substrate:catalyst was
Kinetic Studies of Bicyclization. The kinetics of the
aminodiolefin bicyclization 15 f 16 were investigated by in
situ 1H NMR. With constant catalyst concentration, the
conversion of a 40-50-fold molar excess of N-allyl-4-penten-
1-amine mediated by Cp′2SmCH(SiMe3)2 was monitored over
>3 half-lives. The disappearance of the allyl olefinic 1H
resonances (CH2d, δ ∼5.2 ppm) was normalized to CH2-
(SiMe3)2 as an internal standard. The kinetic data (Figure 2)
reveal a linear dependence of aminodialkene concentration on
reaction time over a ∼10-fold concentration range, which
indicates an essentially zero-order dependence of the catalytic
rate on substrate concentration, in analogy to organolanthanide-
catalyzed aminoolefin6b,c and aminoalkyne cyclization.7 This
argues for a similar turnover-limiting step involving intramo-
lecular unsaturated carbon-carbon bond insertion into the
Ln-N bond. When the initial substrate concentration is held
constant and the catalyst precursor concentration is varied over
a 10-fold range (Figure 3), a plot of reaction rate vs precatalyst
concentration indicates the reaction to be the first-order in
catalyst. The empirical rate law is thus given by eq 9, and is
identical to that for organo-f-element catalyzed amino-
olefin6b,c and aminoalkyne cyclization.7a,b The derived rate
employed in C6D6 solution with Cp′2SmCH(SiMe3)2 and the
process monitored by 1H NMR. Intermolecular amination of a
secondary amine having a pendant terminal alkyne generates a
metal-alkenyl species as a presumed intermediate, which then
undergoes intramolecular insertion of the pendant terminal
alkyne moiety to yield an exomethylene dihydropyrrole deriva-
tive (Scheme 3). This product subsequently undergoes isomer-
ization to the more stable aromatic pyrrole skeleton.9,10,28 In
1H NMR monitoring of the reaction, exomethylene resonances

1768 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Li and Marks
Table 3. Intermolecular Hydroamination/Cyclization Results
at δ 6.0-5.2 initially grow in and then gradually disappear
during the catalytic turnover, while aromatic pyrrole-type signals
at δ 6.3-5.8 evolve as the final product. This conversion is
clean and quantitative under catalytic conditions. GC/MS and
NMR of the final reaction mixtures indicate the presence of
only cyclized pyrrole derivative, C6D6, and the expected quantity
of CH2(SiMe3)2. The products were characterized by GC/MS,
1
HRMS and H, 13C, DEPT, 2D, and NOE difference NMR
spectroscopy.
Although more sterically encumbered secondary amines
generally exhibit lower turnover frequencies than primary
amines in catalytic aminoolefin and aminoalkyne hydroamina-
tion/cyclization,6c,7b it can be seen in Table 3 that the secondary
propargylamines exhibit substantial cyclization rates compared
to analogous intramolecular hydroamination/cyclizations6c and
intermolecular primary amine hydroaminations.6a For example,
the turnover frequency for intermolecular conversion of N-
allylpropargylamine (25) to N-allyl-2-(1′-allyl-3′-methyl-
pyrrolyl)methylamine (26) can be as high as 236 h-1 at 60 °C
using Cp′2Sm- as the catalyst (entry 1, in Table 3). More
interestingly, and in marked contrast to the aforementioned
intramolecular secondary amine hydroamination/bicyclizations
(entries 8-12, in Table 1), N-propargyl-4-penten-1-amine (29
in Table 3) preferentially undergoes intermolecular hydro-
amination/cyclization rather than intramolecular hydroamination/
bicyclization or hydroamination/cyclization (29 f 30, in Table
3, eq 11), arguing for a more rapid intermolecular insertion of
the terminal alkyne moiety into the Ln-N bond than intramo-
lecular olefin insertion (intramolecular CtC insertion would
yield a highly strained three-membered ring).
a Cp′2SmCH(SiMe3)2 precatalyst in C6D6. b Me2SiCp′′2NdCH(SiMe3)2
precatalyst in C6D6. c Yield determined by 1H NMR and GC/MS after
vacuum transfer of volatile products. d Isolated yield. e See text.
Scheme 4. Proposed Pathway for Organolanthanide-
Catalyzed Inter- and Intramolecular Hydroamination/
Tricyclization
The regiospecific, tandem C-N and C-C coupling of
N-allylpropargylamine (25 f 26 in Table 3) is unprecedented,
and the proposed reaction sequence is shown in eq 12. The
Ln-C bond initially generated via intermolecular insertion of
the CtC bond into the Ln-N functionality preferentially
undergoes regiospecific terminal alkyne insertion rather than
alkene insertion to form a five-membered pyrrole structure. The
greater insertive reactivity of Ln-N groups with respect to
alkynes than to alkenes was noted above in intramolecular
hydroamination/ bicyclizations (9 f 10, 11 f 12 in Table 1).
Additionally, N-allyl-2-(1′-allyl-3′-methylpyrrolyl)methyl-
amine (26), produced by organolanthanide-mediated intermo-
lecular hydroamination followed by intramolecular cyclization
(Scheme 4, steps i and ii), undergoes subsequent activation on
longer reaction times (Scheme 4, step iii), intramolecular
hydroamination/bicyclization (Scheme 4, step iv), tricyclization
(Scheme 4, step v), and finally protonolysis (Scheme 4, step
vi) to generate a tricyclic polyheterocycle (31). Furthermore,
tricyclization beginning with N-allylpropargylamine (25, Table
(28) For a discussion of such thermal [1, j] sigmatropic rearrangements,
see: (a) March, J. AdVanced Organic Chemistry, 4th ed.; J. Wiley & Sons:
New York, 1992; pp 1121-1125 and references therein. (b) Boger, D. L.;
Ishizak, T.; Kitos, P. A.; Suntornwak, O. J. Org. Chem. 1990, 55, 5823-
5832.

Catalytic Tandem C-N and C-C Bond-Forming Reactions
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1769
3) or 26 (Table 3) is regiospecific, yielding exclusively a
structure assigned to the trans isomer by 2D and NOE difference
NMR spectroscopy. Attempts to intermolecularly couple a
simple terminal alkyne such as 1-hexyne with substituted
propargylamines followed by hydroamination/cyclization were
unsuccessful, yielding only acyclic coupling products (32) and
homodimers (entry 5, Table 3).
monocyclic products. For example, N-allyl-4-pentyn-1-amine
(5) could potentially form a pyrrolizidine skeleton; however,
the second ring-closing reaction is unrealized, presumably due
to intervention of rapid Ln-C protonolysis of the intermediate
by a free or bound substrate containing an acidic N-H and/or
tC-H moiety (Scheme 1, step ii). Although N-allyl-5-
(trimethylsilyl)-4-pentyn-1-amine (7) also contains the elements
of a pyrrolizidine skeleton, a combination of what are apparently
electronic33 and steric impediments prohibits the second (ole-
finic) insertion into the R-(silylvinyl)lanthanide linkage in the
Ln-C bonded intermediate (Scheme 1, step iii). In the case of
second rings greater than five-membered (Table 1, entries 5, 6,
and 12), apparently the greater steric encumbrance and greater
entropic loss32 in the formation of cyclization transition states
renders the second ring closure uncompetitive with rapid Ln-C
bond protonolysis.
With regard to reaction rate, the present bicyclizations are
kinetically comparable to the corresponding aminoalkene6b,c and
aminoalkyne7a,b monocyclizations for the same catalysts, tem-
peratures, and reaction conditions. Furthermore, insertion in
the first ring-closing reaction largely appears to govern the
overall bicyclization rate, and bicyclizations involving alkyne
insertions in the initial ring formation are more rapid than the
corresponding alkene transformations. As noted above, an
appealing explanation for such insertion sequences and rate
effects invokes a sterically more approachable, electron-rich,
and cylindrical CtC π system, a stereoelectronically influenced
four-center transition state (B; electron-withdrawing groups in
the terminal alkyne moieties (R) stabilize; electron-donating
groups in the terminal position destabilize), more nucleophilic
alkynyl sp hybridized carbon atoms,25 as well as a more
exothermic and exergonic alkyne insertion process.22-24
Discussion
The present study reveals efficient, sequential coupling of
organolanthanide-mediated carbon-nitrogen and carbon-
carbon bond formation/cyclization processes to constitute
catalytic cycles, and considerably extends the scope of known
organolanthanide-mediated olefin and alkyne hydroamination/
cyclization, as well as diolefin, dialkyne, and enyne cyclization
processes. In the ensuing discussion, we focus on factors
affecting the course and rates of both inter- and intramolecular
catalytic variants of these reactions, and interpret them in terms
of likely mechanistic patterns.
Intramolecular Hydroamination/Bicyclization. Kinetics
and Mechanism. The results in Table 1 illustrate that both
Cp′2LnCH(SiMe3)2 and Me2SiCp′′2LnCH(SiMe3)2 complexes
are competent precatalysts for the hydroamination/bicyclization
of aminodialkenes, aminodialkynes, and aminoalkenynes. In
all in situ NMR-scale reactions, instantaneous formation of CH2-
(SiMe3)2 from protonolytic reaction of the precatalysts with the
amine substrates is observed, and the concentration remains
constant throughout the course of the catalytic reaction. It can
be seen in Table 2 that variation of Ln3+ ionic radius
significantly affects the reaction rate of the 1 f 2 transformation.
When catalyst and substrate concentration are held constant at
room temperature, the 1 f 2 turnover frequencies increase with
increasing eight-coordinate Ln3+ ionic radius26 (Figure 3).
However, as mentioned in the Results Section, high amine
concentrations and the more “open” Me2SiCp′′2LnCH(SiMe3)2
precatalyst, along with high reaction temperatures can result in
monocyclic byproducts (3-15%), presumably via competing
intermolecular protonolysis of Ln-C intermediates (Scheme 1,
step ii). The more “open” catalysts and high amine concentra-
tions likely enhance amine coordination at the lanthanide center,
thus facilitating Ln-C protonolysis.
In regard to substrates, the present intramolecular catalytic
coupling process is capable of regiospecifically assembling
aminodialkenes, aminodialkynes, and aminoalkenynes in a single
catalytic cycle to generate pyrrolizidine and indolizidine skel-
etons.8,10 Note in Table 1 that alkyne, alkene (entries 1 and 2);
alkyne, alkyne (entry 7); alkene, alkene (entries 8 and 9); and
alkene, alkyne (entry 11) substrates all undergo organo-
lanthanide-mediated sequential bicyclization as exemplified by
Scheme 2. However, only regiospecific five-membered cy-
clization is observed in significant yields for the second ring-
closing reaction. The diminished activity for six- and seven-
membered second ring bicyclitive closure likely reflects a
sterically controlled process5a,29-32 in competion with Ln-C
protonolysis (Scheme 1, steps ii and iii) which leads to
1
In regard to reaction mechanism, the H NMR monitoring
results indicate that intermolecular substrate protonolysis of the
catalyst hydrocarbyl precursors, similar to the case of organo-
lanthanide-catalyzed aminoalkene6b,c and aminoalkyne
cyclization,7a,b rapidly forms CH2(SiMe3)2 and a catalytically
active LnNRR′(NHRR′)x species. As evidenced by the 1H NMR
of the paramagnetic catalysts, this complex undergoes rapid
exchange of bound amine-amide ligands and/or of bound and
free amines6b,c (eq 13). Next, as examplified by in Scheme 2,
the insertion of an unsaturated carbon-carbon moiety into the
Ln-N amido bond yields a five-membered ring. This insertion
is expected to be more rapid and exothermic than the formation
of a strained three-membered ring.
Cp′2LnNRR′(HNRR′)x h
Cp′2LnNRR′(HNRR′)x-1 + HNRR′ (13)
(29) For reviews of the cyclization of acyclic molecules, see; (a)
Nakagaki, R.; Sakuragi, H.; Mutai, K. J. Phys. Org. Chem. 1989, 2, 187-
204. (b) Mandolini, L. AdV. Phys. Org. Chem. 1986, 22, 1-111.
(30) For a review of the cyclization and conformation of hydrocarbon
chains, see: Winnik, M. A. Chem. ReV. 1981, 81, 491-524.
(31) For a review of steric and electronic effects in heterocyclic ring
closures, see: Valters, R. Russ. Chem. ReV. 1982, 51, 788-801.
(32) For discussions of entropic and enthalpic factors in ring-closing
reactions, see: (a) Mandolini, L. Bull. Soc. Chim. Fr. 1988, 2, 173-176.
(b) DeTar, D. F.; Luthra, N. P. J. Am. Chem. Soc. 1980, 102, 4505-4512.
(c) Menger, F. M. Acc. Chem. Res. 1985, 18, 128-134.
The process of olefin insertion into the Ln-C bond generated
in the initial ring-forming amination/cyclization undoubtedly
occurs via a concerted four-center transition state. Such
processes generally have very low activation barriers, as
(33) For discussions of unusual aspects of d0 metal-R-silylvinyl electronic
structure and bonding, see: (a) Horton, A. D.; Orpen, A. G. Organometallics
1991, 10, 3910-3918. (b) Koga, N.; Morokuma, K. J. Am. Chem. Soc.
1988, 110, 108-112. (c) Eisch, J. J.; Piotrowski, A. M.; Brownstein, S.
K.; Gabe, E. J.; Lee, F. L. J. Am. Chem. Soc. 1985, 107, 7219-7220.

1770 J. Am. Chem. Soc., Vol. 120, No. 8, 1998
Li and Marks
suggested by theoretical calculations,34 and numerous examples
of such insertions,3 including spectroscopic3c and X-ray crystal
structure35 studies designed to model such insertions. The
results of the present bicyclizations are consistent with Scheme
1, involving a rapid second CdC/CtC insertion (Scheme 1,
step iii) into the resulting Ln-C bond formed via previous
carbon-carbon multiple bond insertion into the Ln-N bond
(Scheme 1, step i).
dented thermal isomerization to conjugated pyrrole derivatives.7,8
Such catalytic cycles proceed via rather different mechanisms
than other lanthanide- and actinide-mediated terminal alkyne
dimerizations or oligomerizations.4 In those processes, the
plausible pathway involves protonolysis of the metal-alkyl
precursor with simple terminal alkynes to yield [Ln-CtCR]n
species which then undergo subsequent alkyne insertion into
the Ln-CtCRbond affording Ln-alkenyl complexes. These
species then undergo protonolysis by an incoming alkyne,
yielding an acetylene dimer and regenerating the Ln-alkynyl
complex.
In the present catalytic processes, the substrate contains a
Lewis basic/acidic amine group,37 arguing that rapid protonolysis
by N-H functionalities limits the lifetimes/effective concentra-
tions of any Ln-CtCCH2NHR species, as shown in Scheme
3. Intra- or intermolecular protonolysis of Ln-N(R)CH2CtCH
complexes by a tCH moiety to yield a Ln-CtCCH2NHR
complex is estimated from thermochemical data22-24 to be
exothermic by ∼7 kcal/mol, however this pathway is apparently
not competitive with the far more exothermic insertion of alkyne
into the Ln-N bond (∆H ≈ -35 kcal/mol). It is therefore not
surprising that significant quantities of alkyne dimers or trimers
formed via alkyne insertion into metal-carbon bonds4 are not
observed.
The substantial turnover frequencies in the present dimer-
ization process are especially noteworthy compared to more
conventional organolanthanide-catalyzed intermolecular hydro-
amination processes (amine + olefin; amine + alkyne).6a For
example, the secondary amines in the present process are rather
sterically encumbered for efficient intermolecular insertion of
carbon-carbon multiple bonds into the Ln-N functionalities.
However, it is likely that amine substrates bearing pendant
alkyne functionalities are involved in strong acid-base, catalyst-
substrate preorganization, which would facilitate subsequent
insertion of the proximate alkyne moiety into the Ln-N bond
(e.g., D, Scheme 3). Similar LnNHR(NH2R)x amine-amido
adducts38 are detected in chiral organolanthanide systems6b via
1H NMR, and in achiral catalyst systems by both 1H NMR and
The kinetic results for the Cp′2Sm-catalyzed 15 f 16
transformation indicate zero-order behavior in [substrate] and
first-order behavior in [Sm] (eq 9) in analogy to organo-f-
element-mediated intramolecular aminoolefin,6b,c and
aminoalkyne7a,b hydroamination/cyclization. This implies a
similar intramolecular insertive process during the first ring-
closing reaction (carbon-carbon multiple bond insertion into
the Ln-N bond) arguably as the turnover-limiting step. Further
support for such a turnover-limiting insertion scenario is found
in the pronounced correlation of 1 f 2 turnover frequencies
(at constant [Cp′2Ln], [substrate], and temperature) with de-
creasing eight-coordinate Ln3+ ionic radius26 [Ln (ionic radius,
Å), Nt]: La (1.16), 148; Nd (1.11), 45; Sm (1.08), 17; Lu
(0.977), <0.2 h-1) (Figures 2 and 3). This tendency of large
Ln3+ ions to accelerate unsaturated carbon-carbon bond inser-
tion is consistent with the known patterns in organolanthanide-
mediated olefin hydrogenation,3j,6b,12b,c,36 polymerization,12a,c,36f
hydrophosphination,6d hydrosilylation,3d,e and hydroamination.6
Further evidence for this turnover-limiting insertion process
derives from bicyclization turnover frequencies. For example,
in the case of N-allyl-4-hexyn-1-amine (3) vs N-allyl-4-penten-
1-amine (15) (Table 1), which have sterically similar structures,
the former alkyne insertion into the Ln-N bond exhibits a far
greater Nt (777 h-1) than the latter alkene insertion (Nt ) 55
h-1).
Organolanthanide-Catalyzed Intermolecular Sequential
C-N and C-C Bond Formation. The organolanthanides
Cp′2LnCH(SiMe3)2 and Me2SiCp′′2LnCH(SiMe3)2 also serve as
effective catalyst precursors for rapid, coupled intermolecular
C-N + intramolecular C-C bond formation sequences to yield
pyrrole derivatives in a single catalytic cycle. The principal
reaction mechanism for the present intermolecular hydro-
amination/cyclization likely appears to be the pathway previ-
ously proposed for both intermolecular hydroamination6a and
intramolecular hydroamination/bicyclization.3a This would
involve rapid protonolysis of the precatalysts by the amine
substrates to form LnNRR′(NHRR′)x complexes. These sub-
sequently undergo (presumably) turnover-limiting intermolecular
alkyne insertion into the Ln-N bonds (a similar turnover-limiting
insertion was identified in related intermolecular hydroamination6a)
followed by a second rapid intramolecular alkyne insertion into
the resulting Ln-C bond (Scheme 3; the same type of insertion
was observed in intramolecular cyclizations3a,4b-g). Subsequent
rapid intra- or intermolecular protonolysis of the Ln-C bonds
generates exo-methylene heterocycles which undergo prece-
X-ray crystal structure studies.6c In the present intermolecular
hydroamination/cyclization process, the formation of amine-
(37) For a recent review of equilibrium acidities of organic compounds,
see: (a) Bordwell, F. G.; Zhang, X.-M. Acc. Chem. Res. 1993, 26, 510-
517 and references therein. The pKa value for NH3 is 32.5,37b that for
HCtCH, 24.37c (b) Perrin, D. D. Ionisation Constants of Inorganic Acids
and Bases in Aqueous Solution, 2nd ed.; Pergamon Press: Oxford, U.K.,
1982. (c) Wooding, N. S.; Higginson, W. E. C. J. Chem. Soc. 1952, 774.
(38) For related examples, see: (a) Hogerheide, M. P.; Boersma, J.; Van
koten, G. Coord. Chem. ReV. 1996, 155, 87-126. (b) Deelman, B. J.; Booij,
M.; Meetsma, A.; Teuben, J. H. Kooijman, H.; Spek, A. L. Organometallics
1995, 14, 2306-2317. (c) Evans, W. J.; Keyer, R. A.; Rabe, G. W.;
Drummond, D. K.; Ziller, J. W. Organometallics 1993, 12, 4664-4667.
(d) Qian, C.; Zhu, D. J. Organomet. Chem. 1993, 445, 79-84. (e)
Schumann, H.; Meese-Marktscheffel, J. A.; Dietrich, A.; Gorlitz, F. H. J.
Organomet. Chem. 1992, 430, 299-315. (f) Schumann, H.; Loebel, J.;
Pickardt, J.; Qian, C.; Xie, Z. Organometallics 1991, 10, 215-219. (g)
Shapiro, P. J.; Henling, L. M.; Marsh, R. E.; Bercaw, J. E. Inorg. Chem.
1990, 29, 4560-4565. (h) den Haan, K. H.; Wielstra, Y.; Teuben, J. H.
Organometallics 1987, 6, 2053-2060. (i) Watson, P. L. J. Chem. Soc.,
Chem. Commun. 1983, 276-277. (j) Watson, P. L. J. Am. Chem. Soc. 1982,
104, 337-339.
(34) (a) Yoshida, N.; Koga, N.; Morokuma, K. Organometallics 1995,
14, 746-758 and references therein. (b) Kawamura-Kuribayashi, H.; Koga,
N.; Morokuma, K. J. Am. Chem. Soc. 1992, 117, 2359-2366.
(35) Wu, Z.; Jordan, R. F.; Petersen, J. L. J. Am. Chem. Soc. 1995, 117,
5868-5868.
(36) (a) Conticello, V. P.; Brard, L.; Giardello, M. A.; Tsuji, T.; Sabat,
M.; Stern, C.; Marks, T. J. J. Am. Chem. Soc. 1992, 114, 2761-2762. (b)
Molander, G. A.; Hoberg, J. O. J. Org. Chem. 1992, 57, 3266-3268. (c)
Fendrick, C. M.; Schertz, L. D.; Day, V. W.; Marks, T. J. Organometallics
1988, 7, 1828-1838. (d) Heeres, H. J.; Renkema, J.; Booji, M.; Meetsma,
A.; Teuben, J. H. Organometallics 1988, 7, 2495-2502. (e) den Haan, K.
H.; de Boer, J. L.; Teuben, J. H.; Spek, A. L.; Kajic-Prodic, B.; Hays, G.
R.; Huis, R. Organometallics 1986, 5, 1726-1733. (f) Mauermann, H.;
Marks, T. J. Organometallics 1985, 4, 200-202.

Catalytic Tandem C-N and C-C Bond-Forming Reactions
J. Am. Chem. Soc., Vol. 120, No. 8, 1998 1771
amido adducts is expected to readily place the attached alkyne
moieties in favorable geometries within the lanthanide coordina-
tion spheres for facile alkyne insertion into the Ln-N bonds. It
is therefore not surprising that such intermolecular insertion
processes can compete favorably with intramolecular insertion
(eq 11). That products resulting from simple alkyne insertion
into an Ln-N bond are not detected in the organolanthanide-
mediated intermolecular reaction of N-allylpropargylamine with
excess 1-hexyne (Table 3, entry 5) also supports the proposed
catalyst-substrate preinteraction.
tricyclization (Scheme 4, step v) leading to the unique fused
pyrrole-pyrazine-pyrrole skeleton.9,10f-i
The present intermolecular C-N + intramolecular C-C bond
fusion processes represent a direct and efficient method for
catalytically generating unusual amino-pyrrole derivatives from
simple substituted propargylamines. The apparently straight-
forward reaction mechanism, extendible reaction scope, substrate
versatility, as well as the regiospecific and stereospecific product
formation represents a potentially useful approach to important
heterocycle skeletons such as pyrrole and pyrazine derivatives.
The observation that Ln-C bonds generated by intermolecu-
lar amination undergo cyclization with the pendent alkyne
moieties rather than intra- or intermolecular amine protonolysis
is in accord with the aforementioned rapid carbon-carbon bond
formation processes as demonstrated in the intramolecular
bicyclization. Furthermore, this insertion is favored for alkynes
over alkenes (eq 12), and is analogous to the aforementioned
intra- and intermolecular additions of Ln-N moieties to
carbon-carbon multiple bonds, reflecting the aforementioned
preferential alkyne nucleophilic addition.22,24,25 That the only
observed products are those expected on the basis of sequential
C-N, C-C coupling, strongly supports such a cyclization
mechanistic scenario. However, note that this intermolecular
hydroamination/cyclization can be followed by subsequent
sequential intramolecular hydroamination/cyclization to generate
a tricyclic heterocycle via the aforementioned bicyclization
process (Scheme 4). Also noteworthy here is the stereochemical
course of the tricyclization process. Only the trans-isomer is
observed in the sequential intermolecular C-N, intramolecular
C-C, C-N, and C-C bond formation process. This result
argues for substantial steric control as well as intramolecular
amine coordination to the lanthanide center during the intra-
molecular amination/bicyclization (Scheme 4, step iv) and
Conclusions
Organolanthanide centers can mediate unusual tandem se-
quences of insertive C-N and C-C bond-forming processes
in intramolecular as well as intermolecular hydroamination/
cyclization, and such transformations can be readily integrated
into novel and regioselective catalytic cycles with high turnover
frequencies. Of note is the attraction of assembling pyrrolizi-
dine, indolizidine, and pyrrole-pyrazine-pyrrole skeletons
having varying degrees of unsaturation and substitutable groups,
hence points for subsequent functionalization, in a single
catalytic cycle. Significantly, the catalyst-substrate interaction
via adduct formation can be applied to intermolecular catalytic
reactions, resulting in preferential intermolecular insertion over
intramolecular insertion. Followed by intramolecular cycliza-
tion, this provides an efficient route to pyrrole and pyrazine
derivatives. Additional applications are currently under inves-
tigation.
Acknowledgment. We thank the National Science Founda-
tion for support of this research under Grant CHE-961889. We
thank Dr. I. D. L. Albert for the AM-1 calculations.
JA972643T