Calcium-mediated intramolecular hydroamination catalysis

Article abstract of DOI:10.1021/ja043576n

The calcium-catalyzed intramolecular hydroamination of alkenes and alkynes is reported. The β-diketiminato complex [{HC(C(Me)₂N-2,6-ⁱPr₂C₆H₃)₂}-Ca{N(SiMe₃)₂}(THF)] affects catalytic cyclization of a range of aminoalkenes and aminoalkynes with activities that are broadly commensurate to those of established rare earth catalysts. Copyright

Full text of DOI:10.1021/ja043576n

Published on Web 02/01/2005
Calcium-Mediated Intramolecular Hydroamination Catalysis
Mark R. Crimmin, Ian J. Casely, and Michael S. Hill*
Department of Chemistry, Imperial College London, Exhibition Road, South Kensington, London, SW7 2AZ, UK
Received October 22, 2004; E-mail: mike.hill@imperial.ac.uk
Scheme 1
Hydroamination, the formal addition of an N-H bond across
C-C unsaturation, is an attractive and atom-efficient route to
nitrogen-containing molecules. The widespread search for homo-
geneous reagents capable of catalyzing these processes has begat
a burgeoning and mechanistically diverse area of research that has
utilized alkali metal, transition element, and 5f and 4f element
complexes.1 This latter class of catalyst, devised and pioneered by
Marks and co-workers, is especially notable for its highly efficient
catalytic cyclization of aminoalkenes, aminoalkynes, aminoallenes,
and aminodienes.2 The availability of a single common oxidation
state (+3) precludes oxidative-addition/reductive-elimination as a
viable mechanistic pathway for organolanthanide catalysts. Rather,
reactivity patterns are defined by alkene insertion into polar (largely
ionic) Ln-X (X ) C, N) bonds and σ-bond metathesis, both of
which may be modulated by judicious selection of Ln3+ cationic
radius and supporting ligation.3
mixing the components in C6D6 to yield a 1H NMR spectrum
displaying a series of resonances analogous to literature data for
the expected 2-ethylpyrroline cyclization product, Ia, but shifted
upfield by ca. 0.45-0.50 ppm. A repeat of this stoichiometric
reaction on a preparative scale at room temperature, followed by
hydrolytic workup and analysis of the organic-soluble products by
1H NMR and GCMS, provided irrefutable evidence of hydroami-
nation and cyclization. 2-Ethylpyrroline was identified as the sole
organic product by comparison to literature chemical shift data as
well as its characteristic fragmentation pattern under chemical
ionisation.11
We then investigated the more challenging intramolecular
hydroamination of 1-aminopent-4-ene, II. An NMR scale reaction
between stoichiometric quantities of substrate II and 1 resulted in
complete consumption of the aminoalkene within 24 h at room
temperature and the appearance of several new resonances remi-
niscent of the expected 2-methylpyrrolidine product, IIa. A
limitation of our chosen catalytic system was also apparent since
the reaction proceeded alongside redistribution to the known
homoleptic calcium species [{HC(C(Me)2N-2,6-iPr2C6H3)2}2Ca],
2.6,12 This behavior is consistent with our previous studies of the
reactivity of 1 with primary amines, where we have noted that even
a minor reduction in the overall steric demands of the ligand
substituents can result in irreversible (Schlenk-type) redistribution
to the homoleptic species.7
Livinghouse has reported that homoleptic yttrium and lanthanide
bis(trimethylsilyl)amides [Ln{N(SiMe3)2}3] are competent catalysts
for the intramolecular hydroamination of aminoalkenes.13 We
therefore attempted an NMR scale hydroamination of II with a 10
mol % catalyst loading of the simple homoleptic calcium bis-
(trimethylsilyl)amide [Ca{N(SiMe3)2}2(THF)2], 3. Although this
reaction resulted in complete protonolysis of the bis(trimethylsilyl)-
amide ligands, there was no sign of alkene hydroamination, even
after extended periods at 60 °C.14 In contrast, a similar catalytic
reaction (10 mol % catalyst loading) between II and the â-diketimi-
nato supported complex 1 resulted in the complete conversion of
the aminoalkene to 2-methylpyrrolidine, IIa, over a 21 h period at
25 °C. Although catalytic turnover was also accompanied by
The heavier group 2 metals Ca, Sr, and Ba constitute a relatively
underexploited, yet inexpensive and benign, series of elements that,
as a result of their large radii and electropositive character, possess
a number of features common to the 4f series.4 The present
submission describes our initial efforts to define a catalytic,
molecular hydroamination process based upon a well-defined
calcium coordination complex. In this respect it is notable that
Harder has demonstrated previously that olefinic C-C insertion
into calcium benzyl complexes provides a viable route to polysty-
rene and that a degree of stereoregularity may result from the
application of a kinetically stabilizing fluorenyl-based co-ligand.5
A speculative mechanism for the intramolecular hydroamination
of a generalized R,ω-aminoalkene catalyzed by the synthetically
convenient â-diketiminato calcium bis(trimethylsilyl)amide [{HC-
(C(Me)2N-2,6-iPr2C6H3)2}Ca-{N(SiMe3)2}(THF)], 1,6 is presented
in Scheme 1. Although this mechanism is directly analogous to
that elucidated for 4f element-mediated cyclization,2 it displays
several features with only limited precedent in calcium amide
chemistry. We have shown previously that stoichiometric trans-
amination of the bis(trimethylsilyl)amide substituent of 1 (step A
of Scheme 1) occurs readily to produce a variety of heteroleptic
calcium primary amides and anilides.7 Although insertion of
heterocumulenes into Ca-X (X ) N, O) bonds is well-documented8
and Westerhausen has reported a unique organobarium complex
formed by insertion of diphenylbutadiyne into a barium phos-
phanide,9 no clear precedent exists for insertion of C-C (alkene
or alkyne) unsaturation into a Ca-N bond. In common with
previously reported lanthanide derivatives, this process (Step B of
Scheme 1) is likely to be rate-limiting for any putative calcium-
mediated hydroamination process.
A stoichiometric reaction of 1 and 1-aminohex-4-yne, I, was
studied on an NMR scale to provide an initial assessment of the
viability of an intramolecular alkyne insertion process. Our choice
of an internal alkyne was dictated by previous reports that calcium
bis(trimethylsilyl)amides are sufficiently basic to effect deproto-
nation of terminal alkynes.10 I was consumed within 30 min of
9
2042
J. AM. CHEM. SOC. 2005, 127, 2042-2043
10.1021/ja043576n CCC: $30.25 © 2005 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Catalytic Reactivity of Aminoalkenes and 1
IV, V, and VI also proceeds to ca. 90% conversion in C6D6 under
the NMR conditions listed in Table 1 with catalyst loadings as low
as 2%. Rapid conversion was also observed in the same reactions
performed on a preparative scale. The heterocyclic products IIIa,
IVa, Va,b, and VIa were isolated in ca. 70% yield and characterized
1
by comparison of their H NMR spectra to literature data (see
Supporting Information).
Although the borderline configurational stability of 1 limits
accurate evaluation of substrate turnover, the catalytic activities
reported herein are broadly commensurate with those achieved
recently by Piers and Schafer with the cationic scandium species
[Sc{CH(C(tBu)N-2,6-iPr2C6H3CMe)2}(CH3){CH3B-(C6F5)3}].15 The
highly electrophilic scandium center of this latter complex, which
also features similar â-diketiminato ligation, is formally isoelec-
tronic to the calcium center of 1. Although the utility of 1 may be
limited by its lability to solution exchange equilibria, the low cost
and availability of this alkaline earth metal offers potentially
significant commercial advantages over group 3 and lanthanide-
based methodology. We are continuing to explore the scope of this
reactivity and to address the solution lability of the catalytic alkaline
earth species.
a Entries 1-3, 10 mol % cat. loading. Entry 4, 20 mol % (10 mol %
required 72 h to produce 85% conversion). b Determined by 1H NMR in
C6D6.
competitive redistribution to 2, these observations established
beyond doubt the viability of the catalytic scheme illustrated by
Scheme 1.
Acknowledgment. The Royal Society is thanked for a Univer-
sity Research Fellowship (M.S.H.).
Although the bidentate â-diketiminate ligand provides a sup-
porting environment of only moderate kinetic stability,7b the facile
cyclization of II encouraged further evaluation of the scope of the
catalytic reactivity of 1 with a range of aminoalkene substrates.
These reactions were initially undertaken on an NMR scale and
are summarized in Table 1. In contrast to the reaction with substrate
II, reactions of the geminally substituted 1-aminopent-4-enes III-V
proceeded rapidly at room temperature with no apparent redistribu-
tion to the homoleptic species 2. This is a likely consequence of
the increased rate of reaction (a Thorpe-Ingold effect) as well as
the increased kinetic stability imparted by the geminal aminoalkene
substituents during catalytic turnover. In each case the olefinic
species was consumed within 15 min to produce the target
4-substituted 2-methylpyrrolidines, IIIa and IVa, and the diaster-
eomers Va,b (entries 1-3 of Table 1).
Supporting Information Available: Experimental procedures and
NMR spectra of catalytic aminoalkene hydroamination reactions
(Figures S1-S4). This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Recent general reviews: (a) Togni, G.; Gru¨tzmacher, H. Catalytic
Heterofunctionalisation; VCH: Weinheim, Germany, 2001; p 91. (b)
Mu¨ller, T. E.; Beller, M. Chem. ReV. 1998, 98, 675. (c) Nobis, M.;
Driessen-Ho¨lscher, B. Angew. Chem., Int. Ed. 2001, 40, 3983. (d)
Bytschkov, I.; Doye, S. Chem. Soc. ReV. 2003, 32, 104.
(2) For a recent general overview, see: Hong, S.; Marks, T. J. Acc. Chem.
Res. 2004, 37, 673.
(3) Molander, G.; Romero, J. A. C. Chem. ReV. 2002, 102, 2161.
(4) Greenwood, N. N.; Earnshaw, A. Chemistry of the Elements, 2nd ed.;
Butterworth-Heinemann: Oxford, 1997.
(5) (a) Harder, S.; Feil, F.; Weeber, A. Organometallics 2001, 20, 1044. (b)
Harder, S.; Feil F.; Knoll, K. Angew. Chem., Int. Ed. 2001, 40, 4261.
(6) (a) Chisholm, M. H.; Gallucci, J. C.; Phomphrai, K. Chem. Commun. 2003,
48. (b) Chisholm, M. H.; Gallucci, J. C.; Phomphrai, K. Inorg. Chem.
2004, 43, 6717.
(7) (a) Avent A. G.; Crimmin M. R.; Hill M. S.; Hitchcock P. B. J. Chem.
Soc., Dalton Trans. 2004, 3166. (b) Avent A. G.; Crimmin M. R.; Hill
M. S.; Hitchcock P. B. J. Chem. Soc., Dalton Trans. 2005, 278.
(8) For example: (a) Bezougli, I. K.; Bashall, A.; McPartlin, M.; Mingos, D.
M. P. J. Chem. Soc., Dalton Trans. 1998, 2671. (b) Westerhausen, M.;
Schwartz, W. Z. Naturforsch. 1992, 47b, 453.
A number of common features were apparent in all of these
reactions (Figures S1-S3). Production of bis(trimethylsilyl)amine
was incomplete, and while the relative species ratios were char-
acteristic of the individual reactions, both free and calcium-adducted
heterocyclic products could be identified. A minor quantity of the
calcium primary amide intermediate (produced by step A of Scheme
1) was also apparent from the observation of a heavily shielded
1
and broadened multiplet at ca. -0.7 ppm in the H NMR spectra.
(9) Westerhausen, M.; Digeser, M. H.; No¨th, H.; Seifert, T.; Pfitzner, A. J.
Am. Chem. Soc. 1998, 120, 6722.
These latter resonances are assigned to the proton of the calcium-
bound N-H group by comparison to a similar resonance observed
in the spectrum of the well-defined calcium primary amide [Ca-
{(C(Me)N-2,6-iPr2C6H3CMe)2CH}{µ-NH(CH2)2OMe}]2.7
Although the addition of 1-amino-2,2-dimethylhex-1-ene, sub-
strate VI, to 1 resulted in the instantaneous appearance of the
primary amide intermediate, the desired 6-exo-trig cyclization (entry
4 of Table 1) was too slow to be observed at room temperature.
Heating this sample to 60 °C over a 24 h period, however, resulted
in catalytic production of 2-methyl-5,5-dimethylpiperidine, VIa, the
homoleptic compound 2, and the protonated â-iminoenamine ligand
precursor (Figure S4). The catalytic cyclization of substrates III,
(10) (a) Burkey, D. J.; Hanusa, T. P. Organometallics 1996, 15, 4971. (b)
Chadwick, S.; Englich, U.; Ruhlandt-Senge, K. Inorg. Chem. 1998, 37,
4718.
(11) Although both 2-ethylpyrroline and 1-aminohex-4-yne possess the same
molecular weight, the former is distinguishable by its first daughter ion.
Tehrani, K. A.; Borremans, D.; De Kimpe, N. Tetrahedron 1999, 55, 4133.
(12) Harder, S. Organometallics 2002, 21, 3682.
(13) Kim, Y. K.; Livinghouse, T.; Bercaw, J. E. Tetrahedron Lett. 2001, 42,
2933.
(14) A small amount of oily material, assumed to be the homoleptic calcium
primary amide [Ca{NH(CH2)3CHdCH2}2]n, was also observed to have
deposited from solution.
(15) Lauterwasser, F.; Hayes, P. G.; Bra¨se, S.; Piers, W. E.; Schafer, L.
Organometallics 2004, 23, 2234.
JA043576N
9
J. AM. CHEM. SOC. VOL. 127, NO. 7, 2005 2043