[Ir(COD)Cl]2 as a catalyst precursor for the intramolecular hydroamination of unactivated alkenes with primary amines and secondary alkyl- or arylamines: A combined catalytic, mechanistic, and computational investigation

Article abstract of DOI:10.1021/ja908316n

The successful application of [Ir(COD)Cl]₂ as a precatalyst for the intramolecular addition of primary as well as secondary alkyl- or arylamines to unactivated olefins at relatively low catalyst loading is reported (25 examples), along with a c

Full text of DOI:10.1021/ja908316n

Published on Web 12/09/2009
[Ir(COD)Cl]2 as a Catalyst Precursor for the Intramolecular
Hydroamination of Unactivated Alkenes with Primary Amines
and Secondary Alkyl- or Arylamines: A Combined Catalytic,
Mechanistic, and Computational Investigation
Kevin D. Hesp,† Sven Tobisch,*,‡ and Mark Stradiotto*,†
Department of Chemistry, Dalhousie UniVersity, Halifax, NoVa Scotia B3H 4J3, Canada, and
School of Chemistry, UniVersity of St Andrews, St Andrews, KY16 9ST, United Kingdom
Received September 30, 2009; E-mail: mark.stradiotto@dal.ca; st40@st-andrews.ac.uk
Abstract: The successful application of [Ir(COD)Cl]2 as a precatalyst for the intramolecular addition of
primary as well as secondary alkyl- or arylamines to unactivated olefins at relatively low catalyst loading is
reported (25 examples), along with a comprehensive experimental and computational investigation of the
reaction mechanism. Catalyst optimization studies examining the cyclization of N-benzyl-2,2-diphenylpent-
4-en-1-amine (1a) to the corresponding pyrrolidine (2a) revealed that for reactions conducted at 110 °C
neither the addition of salts (NnBu4Cl, LiOTf, AgBF4, or LiB(C6F5)4 ·2.5OEt2) nor phosphine coligands served
to enhance the catalytic performance of [Ir(COD)Cl]2. In this regard, the rate of intramolecular hydroamination
of 1a employing [Ir(COD)Cl]2/L2 (L2 ) 2-(di-t-butylphosphino)biphenyl) catalyst mixtures exhibited an
inverse-order dependence on L2 at 65 °C, and a zero-order rate dependence on L2 at 110 °C. However,
the use of 5 mol % HNEt3Cl as a cocatalyst was required to promote the cyclization of primary aminoalkene
substrates. Kinetic analysis of the hydroamination of 1a revealed that the reaction rate displays first order
dependence on the concentration of Ir and inverse order dependence with respect to both substrate (1a)
and product (2a) concentrations; a primary kinetic isotope effect (kH/kD ) 3.4(3)) was also observed. Eyring
and Arrhenius analyses for the cyclization of 1a to 2a afforded ∆Hq ) 20.9(3) kcal mol-1, ∆Sq ) -23.1(8)
cal/K·mol, and Ea ) 21.6(3) kcal mol-1, while a Hammett study of related arylaminoalkene substrates
revealed that increased electron density at nitrogen encourages hydroamination (F ) -2.4). Plausible
mechanisms involving either activation of the olefin or the amine functionality have been scrutinized
computationally. An energetically demanding oxidative addition of the amine N-H bond to the IrI center
precludes the latter mechanism and instead activation of the olefin CdC bond prevails, with [Ir(COD)-
Cl(substrate)] M1 representing the catalytically competent compound. Notably, such an olefin activation
mechanism had not previously been documented for Ir-catalyzed alkene hydroamination. The operative
mechanistic scenario involves: (1) smooth and reversible nucleophilic attack of the amine unit on the metal-
coordinated CdC double bond to afford a zwitterionic intermediate; (2) Ir-C bond protonolysis via stepwise
proton transfer from the ammonium unit to the metal and ensuing reductive elimination; and (3) final
irreversible regeneration of M1 through associative cycloamine expulsion by new substrate. DFT unveils
that reductive elimination involving a highly reactive and thus difficult to observe IrIII-hydrido intermediate,
and passing through a highly organized transition state structure, is turnover limiting. The assessed effective
barrier for cyclohydroamination of a prototypical secondary alkylamine agrees well with empirically
determined Eyring parameters.
Introduction
ing the intramolecular hydroamination of unactivated aminoalk-
enes under mild conditions and with broad substrate scope are
The hydroamination of alkenes, alkynes, and related unsatur-
ated substrates represents an attractive strategy for the construc-
tion of nitrogen-containing compounds that circumvents the
formation of byproduct in the creation of a C-N linkage.1 In
particular, intramolecular variants of these reactions are appeal-
ing as atom-economical routes to nitrogen-containing hetero-
cycles that are prevalent in naturally occurring and/or biologi-
cally active molecules.2 Although the synthetic potential of
cyclohydroamination is significant, general methods for promot-
still lacking. A diverse number of catalyst systems are known
to effect such cyclizations,1a including (but not restricted to)
those based on rare earth elements and actinides,3 alkali and
alkaline earth metals,4 Group 4 metals,5 and Brønsted acids;6
nonetheless, the pursuit of increased substrate scope and
functional group tolerance has prompted the development of
alternative catalysts based on metals from Groups 8-11. While
progress has been made in the use of Fe,7 Pd,8 and Au9,10
catalysts for the intramolecular addition of secondary amides,
carbamates, or ureas to unactivated olefins, the identification
of late metal catalysts for analogous transformations involving
† Dalhousie University.
‡ University of St Andrews.
9
10.1021/ja908316n  2010 American Chemical Society
J. AM. CHEM. SOC. 2010, 132, 413–426 413

A R T I C L E S
Hesp et al.
secondary alkyl- or arylamines, as well as primary amines, has
proven more elusive. The first catalyst system of this type was
described in 2005 by Widenhoefer and co-workers,11 who
reported the use of [PtCl2(H2CdCH2)]2/PPh311a and later PtCl2/
biarylphosphine11b for the cyclization of secondary alkylami-
noalkenes, including one example involving a 1,1-disubstituted
olefin. In 2008, Liu and Hartwig12 disclosed the use of
[Rh(COD)2]BF4/Cy-DavePhos (COD ) η4-cyclooctadiene; Cy-
DavePhos ) 2-dicyclohexylphosphino-2′-N,N-dimethylamino-
biphenyl) for the hydroamination of aminoalkenes that feature
primary or secondary alkylamines and terminal or internal
alkenes. In the same year, the use of Rh and Ir complexes
supported by pincer-type N-heterocyclic carbene ligands for the
cyclization of terminal alkenes by pendant secondary alkyl- and
phenylamines was described by Hollis and co-workers.13
Following the appearance of our preliminary report in this area,14
Sawamura and co-workers15 disclosed the use of Cu(OtBu)/
Xantphos (Xantphos ) 4,5-bis(diphenylphosphino)-9,9-dim-
ethylxanthene) for the hydroamination of alkenes by tethered
primary or secondary alkylamines, including an example of a
1,1-disubstituted secondary aminoalkene substrate. Notwith-
standing this recent progress, the identification of late metal
catalysts for use in promoting the intramolecular addition of
primary as well as secondary alkyl- and arylamines to both
terminal and internal olefins remains an important and significant
challenge. Furthermore, with the exception of the NMR
spectroscopic characterization of catalytic intermediates in the
breakthrough Pt catalyst system disclosed by Bender and
Widenhoefer,11a no kinetic or mechanistic data pertaining to
the cyclohydroamination of alkylamino-alkenes are provided in
the ensuing reports.12-15
(1) (a) Mu¨ller, T. E.; Hultzsch, K. C.; Yus, M.; Foubelo, F.; Tada, M.
Chem. ReV. 2008, 108, 3795. (b) Hartwig, J. F. Nature 2008, 455,
314. (c) Widenhoefer, R. A. Chem.sEur. J. 2008, 14, 5382. (d)
Chianese, A. R.; Lee, S. J.; Gagne´, M. R. Angew. Chem., Int. Ed.
2007, 46, 4042. (e) Severin, R.; Doye, S. Chem. Soc. ReV. 2007, 36,
1407. (f) Brunet, J.-J.; Chu, N.-C.; Rodriguez-Zubiri, M. Eur. J. Inorg.
Chem. 2007, 4711. (g) Aillaud, I.; Collin, J.; Hannedouche, J.; Schulz,
E. Dalton Trans. 2007, 5105. (h) Widenhoefer, R. A.; Han, X. Eur.
J. Org. Chem. 2006, 4555. (i) Hultzsch, K. C. AdV. Synth. Catal. 2005,
347, 367. (j) Hultzsch, K. C. Org. Biomol. Chem. 2005, 3, 1819. (k)
Alonso, F.; Beletskaya, I. P.; Yus, M. Chem. ReV. 2004, 104, 3079.
(l) Hartwig, J. F. Pure Appl. Chem. 2004, 76, 507. (m) Roesky, P. W.;
Mu¨ller, T. E. Angew. Chem., Int. Ed. 2003, 42, 2708. (n) Pohlki, F.;
Doye, S. Chem. Soc. ReV. 2003, 32, 104. (o) Nobis, M.; Drieꢀen-
Ho¨lscher, B. Angew. Chem., Int. Ed. 2001, 40, 3983. (p) Mu¨ller, T. E.;
Beller, M. Chem. ReV. 1998, 98, 675.
(2) (a) Hong, S.; Kawaoka, A. M.; Marks, T. J. J. Am. Chem. Soc. 2003,
125, 15878. (b) Trost, B. M.; Tang, W. J. Am. Chem. Soc. 2003, 125,
8744. (c) Molander, G. A.; Dowdy, E. D.; Pack, S. K. J. Org. Chem.
2001, 66, 4344. (d) O’Hagan, D. Nat. Prod. Rep. 2000, 17, 435. (e)
Arredondo, V. M.; Tian, S.; McDonald, F. E.; Marks, T. J. J. Am.
Chem. Soc. 1999, 121, 3633.
(3) For a selection of recent lead references, see: (a) Yuen, H. F.; Marks,
T. J. Organometallics 2009, 28, 2423. (b) Lu, E.; Gan, W.; Chen, Y.
Organometallics 2009, 28, 2318. (c) Aillaud, I.; Collin, J.; Duhayon,
C.; Guillot, R.; Lyubov, D.; Schulz, E.; Trifonov, A. Chem.sEur. J.
2008, 14, 2189. (d) Ge, S.; Meetsma, A.; Hessen, B. Organometallics
2008, 27, 5339. (e) Hannedouche, J.; Aillaud, I.; Collin, J.; Schulz,
E.; Trifonov, A. Chem. Commun. 2008, 3552. (f) Zi, G.; Xiang, L.;
Song, H. Organometallics 2008, 27, 1242. (g) Stubbert, B. D.; Marks,
T. J. J. Am. Chem. Soc. 2007, 129, 4253. (h) Stubbert, B. D.; Marks,
T. J. J. Am. Chem. Soc. 2007, 129, 6149. (i) Vitanova, D. V.; Hampel,
F.; Hultzsch, K. C. J. Organomet. Chem. 2007, 692, 4690. (j) Rasta¨tter,
M.; Zulys, A.; Roesky, P. W. Chem.sEur. J. 2007, 13, 3606. (k)
Riegert, D.; Collin, J.; Daran, J.-C.; Fillebeen, T.; Schulz, E.; Lyubov,
D.; Fukin, G.; Trifonov, A. Eur. J. Inorg. Chem. 2007, 1159. (l) Yu,
X.; Marks, T. J. Organometallics 2007, 26, 365. (m) Rasta¨tter, M.;
Zulys, A.; Roesky, P. W. Chem. Commun. 2006, 874. (n) Gribkov,
D. V.; Hultzsch, K. C.; Hampel, F. J. Am. Chem. Soc. 2006, 128,
3748. (o) Bambirra, S.; Tsurugi, H.; van Leusen, D.; Hessen, B. Dalton
Trans. 2006, 1157. (p) Hultzsch, K. C.; Gribkov, D. V.; Hampel, F.
J. Organomet. Chem. 2005, 690, 4441. (q) Kim, H.; Lee, H. P.;
Livinghouse, T. Chem. Commun. 2005, 5205. (r) Kim, J.-Y.; Living-
house, T. Org. Lett. 2005, 7, 4391. (s) Lauterwasser, F.; Hayes, P. G.;
Bra¨se, S.; Piers, W. E.; Schafer, L. L. Organometallics 2004, 23, 2234.
(t) Hong, S.; Marks, T. J. Acc. Chem. Res. 2004, 37, 673, and
references cited therein.
Encouraged by advances in the Ir-mediated intermolecular
hydroamination of activated alkenes,16-18 as well as recent
progress in the Ir-catalyzed cyclohydroamination of alkynes,16,19
we sought to expand further the synthetic repertoire of late
metal-mediated hydroamination through the identification of
simple Ir precatalysts that exhibit useful substrate scope in the
(8) (a) Cochran, B. M.; Michael, F. E. J. Am. Chem. Soc. 2008, 130, 2786.
(b) Cochran, B. M.; Michael, F. E. Org. Lett. 2008, 10, 329. (c)
Michael, F. E.; Cochran, B. M. J. Am. Chem. Soc. 2006, 128, 4246.
(9) (a) Bender, C. F.; Widenhoefer, R. A. Chem. Commun. 2006, 4143.
(b) Bender, C. F.; Widenhoefer, R. A. Org. Lett. 2006, 8, 5303. (c)
Zhang, J.; Yang, C.-G.; He, C. J. Am. Chem. Soc. 2006, 128, 1798.
(d) Liu, X.-Y.; Li, C.-H.; Che, C.-M. Org. Lett. 2006, 8, 2707. (e)
Han, X.; Widenhoefer, R. A. Angew. Chem., Int. Ed. 2006, 45, 1747.
(10) The Au-catalyzed intermolecular hydroamination of simple R-olefins
with cyclic ureas has been reported recently: Zhang, Z.; Lee, S. D.;
Widenhoefer, R. A. J. Am. Chem. Soc. 2009, 131, 5372.
(4) For some recent references, see: (a) Zhang, W.; Werness, J. B.; Tang,
W. Tetrahedron 2009, 65, 3090. (b) Crimmin, M. R.; Arrowsmith,
M.; Barrett, A. G. M.; Casely, I. J.; Hill, M. S.; Procopiou, P. A.
J. Am. Chem. Soc. 2009, 131, 9670. (c) Arrowsmith, M.; Hill, M. S.;
Kociok-Ko¨hn, G. Organometallics 2009, 28, 1730. (d) Datta, S.;
Gamer, M. T.; Roesky, P. W. Organometallics 2008, 27, 1207. (e)
Martinez, P. H.; Hultzsch, K. C.; Hampel, F. Chem. Commun. 2006,
2221.
(11) (a) Bender, C. F.; Widenhoefer, R. A. J. Am. Chem. Soc. 2005, 127,
1070. (b) Bender, C. F.; Hudson, W. B.; Widenhoefer, R. A.
Organometallics 2008, 27, 2356.
(12) Liu, Z.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130, 1570.
(13) Bauer, E. B.; Andavan, G. T. S.; Hollis, T. K.; Rubio, R. J.; Cho, J.;
Kuchenbeiser, G. R.; Helgert, T. R.; Letko, C. S.; Tham, F. S. Org.
Lett. 2008, 10, 1175.
(5) For a selection of recent lead references, see: (a) Lee, A. V.; Sajitz,
M.; Schafer, L. L. Synthesis 2009, 97. (b) Lian, B.; Spaniol, T. P.;
Horrillo-Martinez, P.; Hultzsch, K. C.; Okuda, J. Eur. J. Inorg. Chem.
2009, 429. (c) Cho, J.; Hollis, T. K.; Helgert, T. R.; Valente, E. J.
Chem. Commun. 2008, 5001. (d) Xiang, L.; Song, H.; Zi, G. Eur.
J. Inorg. Chem. 2008, 1135. (e) Majumder, S.; Odom, A. L.
Organometallics 2008, 27, 1174. (f) Wood, M. C.; Leitch, D. C.;
Yeung, C. S.; Kozak, J. A.; Schafer, L. L. Angew. Chem., Int. Ed.
2007, 46, 354. (g) Lee, A. V.; Schafer, L. L. Eur. J. Inorg. Chem.
2007, 2243, and reference cited therein. (h) Gott, A. L.; Clarke, A. J.;
Clarkson, G. J.; Scott, P. Organometallics 2007, 26, 1729. (i) Mu¨ller,
C.; Loos, C.; Schulenberg, N.; Doye, S. Eur. J. Org. Chem. 2006,
2499. (j) Watson, D. A.; Chiu, M.; Bergman, R. G. Organometallics
2006, 25, 4731. (k) Knight, P. D.; Munslow, I.; O’Shaughnessy, P. N.;
Scott, P. Chem. Commun. 2004, 894.
(14) For our preliminary report of the use of [Ir(COD)Cl]2 as a pre-catalyst
for the intramolecular hydroamination of secondary alkyl- or ary-
lamines (but not primary amines) and unactivated olefins, see: Hesp,
K. D.; Stradiotto, M. Org. Lett. 2009, 11, 1449.
(15) Ohmiya, H.; Moriya, T.; Sawamura, M. Org. Lett. 2009, 11, 2145.
(16) Dorta, R. In Iridium Complexes in Organic Synthesis; Oro, L. A.,
Claver, C., Eds.; Wiley-VCH: Weinheim, 2009; p 145.
(17) (a) Casalnuovo, A. L.; Calabrese, J. C.; Milstein, D. J. Am. Chem.
Soc. 1988, 110, 6738. (b) Dorta, R.; Egli, P.; Zu¨rcher, F.; Togni, A.
J. Am. Chem. Soc. 1997, 119, 10857.
(18) Zhou, J.; Hartwig, J. F. J. Am. Chem. Soc. 2008, 130, 12220.
(19) For some recent reports, see: (a) Field, L. D.; Messerle, B. A.; Vuong,
K. Q.; Turner, P. Dalton Trans. 2009, 3599. (b) Ebrahimi, D.;
Kennedy, D. F.; Messerle, B. A.; Hibbert, D. B. Analyst 2008, 817.
(c) Burling, S.; Field, L. D.; Messerle, B. A.; Rumble, S. L.
Organometallics 2007, 26, 4335. (d) Lai, R.-Y.; Surekha, K.; Hayashi,
A.; Ozawa, F.; Liu, Y.-H.; Peng, S.-M.; Liu, S.-T. Organometallics
2007, 26, 1062. (e) Li, X.; Chianese, A. R.; Vogel, T.; Crabtree, R. H.
Org. Lett. 2005, 7, 5437.
(6) (a) Ackermann, L.; Kaspar, L. T.; Althammer, A. Org. Biomol. Chem.
2007, 5, 1975. (b) Schlummer, B.; Hartwig, J. F. Org. Lett. 2002, 4,
1471.
(7) Komeyama, K.; Morimoto, T.; Takaki, K. Angew. Chem., Int. Ed. 2006,
45, 2938.
9
414 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010

[Ir(COD)Cl]2 as a Catalyst Precursor
A R T I C L E S
Table 1. Screening of Group 9 Catalyst Precursors for the
Intramolecular Hydroamination of 1a to Give 2aa
the [Ir(COD)Cl]2-mediated transformation of 1a into 2a (Ir:1a
) 1:1) were thwarted by the relative complexity of the 1H NMR
spectra obtained under catalytically relevant conditions, it is
worth noting that neither Ir-H nor Ir-D (when using 1a-d1 - the
N-D isotopomer of 1a) resonances were observed over a range
of temperatures (25-80 °C) and throughout no free COD was
entry
catalyst precursor
conversion to 2ab
1-1
1-2
1-3
1-4
1-5
1-6
1-7
1-8
1-9
[Ir(COD)Cl]2
[Ir(COE)2Cl]2
>95
10
7
<5
<5
<5
28c
10c
<5
[Ir(COD)OMe]2
[(η5-C5Me5)IrCl2]2
[Rh(COD)Cl]2
[Ir(COD)2]BF4
[Ir(MeCN)2(COD)]BF4
[Ir(COD)PCy3(py)]PF6
none
2
detected (1H and H NMR).
a Reaction Conditions: 0.25 mmol 1a and 1.0 mol % Ir or Rh in 0.5
mL 1,4-dioxane at 110 °C for 3 h. b On the basis of 1H NMR data
(average of two runs); unless otherwise noted, 2a was observed as the
only reaction product. c Formation of 2a was accompanied by other
byproducts. COD ) η4-1,4-cyclooctadiene; COE ) cyclooctene; Cy )
cyclohexyl; py ) pyridine.
In an effort to further enhance the catalytic performance of
[Ir(COD)Cl]2, the effect of added salts, as well as added
phosphine coligands, was evaluated. In monitoring the trans-
formation of 1a to 2a, the use of 0.25 mol % [Ir(COD)Cl]2 as
a precatalyst alone (110 °C) provided higher rates of conversion
to 2a than were achieved under analogous conditions when using
0.25 mol % [Ir(COD)Cl]2 in the presence of 0.5 mol % NnBu4Cl,
LiOTf, AgBF4, or LiB(C6F5)4 ·2.5OEt2 (Figure S1, Supporting
Information); these results are surprising, given that increased
catalytic activity is often observed for cationic late metal
hydroamination catalysts when large, less-coordinating anions
are employed.1a While for the hydroamination of 1a the use of
[Ir(COD)Cl]2/2 AgBF4 as a precatalyst mixture afforded com-
plete conversion (1H NMR) to 2a after 3 h, the use of
LiB(C6F5)4 ·2.5OEt2 as a cocatalyst under similar conditions also
generated a substantial amount of a byproduct arising from
alkene isomerization within 1a. In keeping with these observa-
tions, conductivity measurements of 1,4-dioxane solutions of
[Ir(COD)Cl]2 carried out under various conditions, including
in the presence or absence of 1a, as well as at room temperature
or following preconditioning under catalytic conditions, provided
no evidence for heterolytic Ir-Cl bond cleavage.
Given that the use of added phosphines has proven effective
in promoting the intramolecular addition of secondary alky-
lamines to unactivated alkenes in reactions mediated by Rh,12
Pt,11 and Cu15 precatalysts, the influence of such coligands on
the catalytic performance of [Ir(COD)Cl]2 in the cyclization of
1a was surveyed. While no reactivity benefits were derived from
the addition of phosphines in reactions conducted at 110 °C,
the use of bulky RPtBu2 phosphines (R ) biaryl or ferrocenyl;
Table S2, Supporting Information) did prove useful in promoting
the formation of 2a under more mild conditions than could be
achieved by use of [Ir(COD)Cl]2 alone. For example, while the
use of 2.5 mol % [Ir(COD)Cl]2 at 65 °C afforded 60%
conversion to 2a in the absence of byproduct over the course
of 24 h, the incorporation of 5 mol % 2-(di-t-butylphosphino)-
biphenyl (L2, Table S2, Supporting Information) as a coligand
in this reaction afforded 2a quantitatively (1H NMR) under
similar reaction conditions. However, we were surprised to find
that the rate of intramolecular hydroamination of 1a employing
[Ir(COD)Cl]2/L2 catalyst mixtures exhibits an inverse-order
dependence on L2 at 65 °C and a zero-order rate dependence
on L2 at 110 °C (Figure S2, Supporting Information). On the
basis of these data, it is unlikely that species featuring an Ir-P
linkage represent important catalytic intermediates in the
observed hydroamination of 1a. Rather, the reactivity benefits
of employing L2 and related bulky phosphines at 65 °C may
be attributable to the ability of such coligands to discourage
the decomposition of catalytically inactive Ir intermediates,
which in turn can serve as a source of catalytically active and
intramolecular hydroamination of unactivated olefins. We
provide herein a complete account of our studies using [Ir-
(COD)Cl]2 as a precatalyst for the intramolecular addition of
primary as well as secondary alkyl- or arylamines to unactivated
olefins.14 Mechanistic investigations of these intramolecular
hydroamination reactions employing a combination of experi-
mental and computational techniques reveal that such transfor-
mations proceed via an olefin activation mechanism that had
previously not been documented for Ir-catalyzed alkene
hydroamination.16-18
Results and Discussion
Catalyst Identification and Optimization. The ability of
simple group 9 complexes to mediate the intramolecular
hydroamination of unactivated alkenes at relatively low catalyst
loadings (1.0 mol % Rh or Ir) was surveyed by using the
cyclization of the secondary alkylamine 1a (eq 1) to the
corresponding pyrrolidine 2a in 1,4-dioxane (110 °C, 3 h) as a
test reaction (Table 1). Under these conditions, the use of
[Ir(COD)Cl]2 as a precatalyst provided clean conversion to 2a
(1H NMR), and monitoring of the reaction revealed the
transformation to be complete after 1 h. By comparison, other
commercially available neutral Group 9 precatalysts performed
poorly under these test conditions, including [Ir(η2-
cyclooctene)2Cl]2, [Ir(COD)(OMe)]2, [(η5-C5Me5)IrCl2]2,
and [Rh(COD)Cl]2 (entries 1-2 to 1-5, respectively);
inferior performance was also observed for the cationic Ir
precatalysts [Ir(COD)2]BF4, [Ir(MeCN)2(COD)]BF4, and
[Ir(COD)PCy3(pyridine)]PF6 (Crabtree’s catalyst) (entries 1-6
to 1-8, respectively). Further investigations revealed that the
quantitative conversion of 1a into 2a can also be achieved by
use of only 0.125 mol % [Ir(COD)Cl]2 in 1,4-dioxane (3 h, 110
°C), or by employing higher Ir loadings at lower temperatures
(2.5 mol % Ir; 7 h, 80 °C). While this reaction was also shown
to proceed in 1,2-dichloroethane, 1,2-dimethoxyethane, or
toluene, 1,4-dioxane proved to be the optimal solvent among
those surveyed and was used for all subsequent catalytic studies
(Table S1, Supporting Information). In keeping with the
observation that the hydroamination of 1a catalyzed by [Ir-
(COD)Cl]2 exhibits an inverse-order rate dependence on the
concentrations of each of 1a and 2a (Vide infra), the conversion
of 1a to 2a was found to be completely inhibited when
employing 0.5 mol % [Ir(COD)Cl]2 at 110 °C in 1,4-dioxane
in the presence of an equivalent of K2CO3 (relative to 1a) or
1.0 mol % 1,8-diazabicyclo(5.4.0)undec-7-ene (DBU). While
efforts to definitively characterize intermediates formed during
9
J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010 415

A R T I C L E S
Hesp et al.
Table 2. Intramolecular Hydroamination of Unactivated Alkenes by
Table 3. Intramolecular Hydroamination of Unactivated Alkenes by
Secondary Alkylamines Employing [Ir(COD)Cl]2 as a Precatalysta
Secondary Arylamines Employing [Ir(COD)Cl]2 as a Precatalysta
a Conditions: 0.25 mmol aminoalkene in 0.5 mL 1,4-dioxane at
110 °C; conversion to product >95% (1H NMR) unless stated. b Isolated
c
percent yield unless stated (average of two runs). 1H NMR percent
d
yield. Conversion to the piperidine was 20% (1H NMR) with the
balance corresponding to an alkene isomerization product. e No reaction
of the aminoalkene was observed.
herein, no unsaturated hydroamination products arising from
ꢀ-hydride elimination were detected.
In keeping with the Cu(OtBu)/Xantphos catalyst system,15
butincontrasttothePtCl2/biarylphosphine11b and[Rh(COD)2]BF4/
biarylphosphine12 catalyst systems, the use of [Ir(COD)Cl]2
under our typical conditions (e.g., Table 2) proved ineffective
in promoting the hydroamination of aminoalkene substrates
lacking backbone substituents that promote cyclization. Fur-
thermore, the N-Boc relative of 1a (i.e., 1f) proved unreactive
(1H NMR) in the presence of [Ir(COD)Cl]2 (Boc ) t-butoxy-
carbonyl; 1.0 mol % Ir; 3 h, 110 °C). However, the observation
that the cyclization of 1a proceeds unhampered (1H NMR) when
using a 1:1 mixture of 1a and 1f confirms that the latter does
not poison the active catalyst derived from [Ir(COD)Cl]2. In
this regard, [Ir(COD)Cl]2 appears to be well-suited as a
precatalyst for conducting chemoselective hydroaminations
involving more complex organic molecules.
Intramolecular Hydroamination of Secondary Arylamines.
The late metal-mediated cyclohydroamination of unactivated
alkenes by secondary arylamines is limited to a single report
featuring the cyclization of two simple N-Ph substrates.13 The
use of [Ir(COD)Cl]2 as a precatalyst established expanded scope
for this class of transformations (Table 3) in enabling the
cyclization the parent N-Ph substrate 3a (entry 3-1), as well
as for the first time arylaminoalkene substrates featuring para
electron-donating (3b and 3c; entries 3-2 and 3-3) and para
electron-withdrawing (3d and 3e; entries 3-4 and 3-5)
substituents. While high isolated yields of 4a-d were obtained
for reactions conducted over the course of 7 h at 110 °C, the
amount of catalyst needed to achieve quantitative conversion
to 4 under these conditions varied significantly. Whereas only
0.25 mol % [Ir(COD)Cl]2 was required to effect the clean
hydroamination of the anisole derivative 3c (entry 3-3), the
use of 2.5 mol % [Ir(COD)Cl]2 afforded only partial conversion
of the trifluoromethyl species 3e to 4e, even after 24 h (entry
3-5); a more detailed kinetic analysis of these reactivity trends
is provided (Vide infra). Our observation that the intramolecular
hydroamination of 3 is significantly enhanced by para electron-
donating substituents (X in eq 2) is divergent from trends
a Conditions: 0.25 mmol aminoalkene in 0.5 mL 1,4-dioxane at
110 °C; conversion to product >95% (1H NMR) unless stated. b Isolated
c
percent yield unless stated (average of two runs). 1H NMR percent
yield. d Diastereomeric ratio (dr, 1H NMR). e Reaction run at 80 °C;
conversion to the piperidine was 75% (1H NMR) with the balance
corresponding to an alkene isomerization product.
phosphine-free Ir species. As such, [Ir(COD)Cl]2 was employed
for all subsequent catalytic studies without added coligands.
Intramolecular Hydroamination of Secondary Alkylamines.
Encouraged by the ability of [Ir(COD)Cl]2 to mediate the
cyclization of 1a, the hydroamination of alternative secondary
alkylaminoalkene substrates, including those featuring polar
groups, was examined (Table 2). The presence of halide (entry
2-2), ester (entry 2-3), or alkoxy (entry 2-4) substituents
within the benzylamines 1b-d did not significantly influence
the progress of the reaction, and the bulky N-methylcyclohexyl
aminoalkene 1e was also cyclized (entry 2-5); in all cases, high
isolated yields were obtained (85-89%) despite the use of
relatively low catalyst loading (1-2.5 mol % Ir). Alternative
gem-disubstituted relatives of 1a also proved to be suitable
hydroamination substrates (entries 2-6 to 2-8), as did the
monosubstituted isopropyl-containing species featured in entry
2-9. The hexenylamine substrate featured in entry 2-10 was
also efficiently cyclized in the presence of [Ir(COD)Cl]2 to afford
the corresponding piperidine derivative. In the pursuit of more
difficult substrates, we turned our attention to the cyclohy-
droamination of unactivated disubstituted alkenes. While the
use of [Ir(COD)Cl]2 proved effective in promoting the cycliza-
tion of gem-disubstituted olefins (entries 2-11 and 2-12),
hydroamination of the internal olefinic substrates featured in
entries 2-13 and 2-14 presented a more significant challenge.
Notably, throughout the course of these and the other [Ir-
(COD)Cl]2-mediated cyclohydroamination reactions reported
9
416 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010

[Ir(COD)Cl]2 as a Catalyst Precursor
A R T I C L E S
that Brønsted acids can act as catalysts6 or cocatalysts20 in
promoting the intramolecular hydroamination of unactivated
alkenes and alkynes provided the motivation to examine the
use of [Ir(COD)Cl]2/Brønsted acid catalyst mixtures. To the best
of our knowledge, the Brønsted acid-mediated intramolecular
hydroamination of unactivated alkenes with primary amines is
limited to a single example by Ackermann and co-workers,6a
in which the cyclization of the substrate featured in entry 4-1
(Table 4) was achieved in 83% yield by use of 20 mol %
[PhMe2NH]B(C6F5)4 (18 h, 120 °C). In the pursuit of a more
convenient and cost-effective cocatalyst, HNEt3Cl and
CF3SO3H6b were each evaluated. While none of 2.5 mol %
[Ir(COD)Cl]2, 10 mol % HNEt3Cl, and 10 mol % CF3SO3H
individually proved capable of promoting the hydroamination
of the primary aminoalkene substrates featured in Table 4 (24
h, 110 °C), the use of 2.5 mol % [Ir(COD)Cl]2/5 mol % HNEt3Cl
as a precatalyst mixture provided excellent conversions to the
targeted N-H containing cyclic secondary amines. Inferior
results were obtained, including the formation of byproducts,
when 5 mol % CF3SO3H was used as a cocatalyst under similar
conditions. Control experiments confirmed that neither 10 mol
% NnBu4Cl nor mixtures comprised of 2.5 mol % [Ir(COD)Cl]2/
5 mol % NnBu4Cl are capable of promoting the cyclization of
these substrates under our experimental conditions. The catalytic
performance of [Ir(COD)Cl]2/HNEt3Cl in the cyclization of these
primary aminoalkene substrates is competitive with that of the
previously reported “two-component” late metal catalysts
[Rh(COD)2]BF4/biarylphosphine12 and Cu(OtBu)/Xantphos.15
Kinetic Studies. A number of distinct mechanistic pathways
have been uncovered for the hydroamination of alkenes.1 In
the case of late metal-mediated transformations these include
(but are not restricted to) processes involving: nucleophilic attack
of an amine on a metal-coordinated alkene,8,11,21 an η3-allyl or
related complex,22 or an η6-vinylarene complex;23 N-H bond
activation followed by alkene insertion into an M-NR2
linkage;16-18 amine coordination followed by proton-transfer
to an activated alkene;24 and nucleophilic attack of a metal
amido species on an activated alkene.25 While experimental
evidence for Ir-mediated N-H bond activation/alkene insertion
reaction sequences has been obtained for the intermolecular
hydroamination of activated alkenes such as norbornene,16-18
mechanistic experiments indicate that the Ir-catalyzed intramo-
lecular hydroamination (and hydroalkoxylation) of alkynes can
proceed via alternative reaction pathways involving alkyne
activation/nucleophilic attack.19e,26 However, a thorough kinetic
analysis of the late metal-mediated cyclohydroamination of
Table 4. Intramolecular Hydroamination of Unactivated Alkenes by
Primary Alkylamines Employing [Ir(COD)Cl]2 and HNEt3Cl (1:2) as
a Precatalyst Mixturea
a Conditions: 0.25 mmol aminoalkene in 0.5 mL 1,4-dioxane at 110
°C; conversion to product >95% (1H NMR) unless stated. b Isolated
c
percent yield unless stated (average of two runs). 1H NMR percent
yield; the remainder is attributed to an alkene isomerization product.
documented by Zhou and Hartwig,18 who reported that both
electron-rich and electron-poor anilines were less reactive than
electron-neutral anilines toward the intermolecular hydroami-
nation of activated bicyclic alkenes catalyzed by (bisphosphi-
ne)Ir species, which was shown to proceed by way of N-H
bond activation/alkene insertion reaction pathway.18 While an
alternative backbone substitution pattern was tolerated (entry
3-6), the hydroamination of a 5-hexenyl substrate proceeded
in low yield (entry 3-7), and efforts to promote the cyclization
of the disubstituted olefinic substrate featured in entry 3-8 by
use of [Ir(COD)Cl]2 were unsuccessful. One might predict
arylamine substrates to be inherently more susceptible to late
metal-mediated hydroamination than alkylamines; the lower
basicity of the former should reduce catalyst inhibition arising
from unproductive substrate and/or product binding, and should
also serve to promote N-H bond cleavage in hydroamination
reactions involving net proton transfer from nitrogen. However,
a comparative reactivity survey revealed that some of the most
effective late metal catalysts known for the intramolecular
hydroamination of unactivated alkenes fail to promote the
cyclization of arylamines such as 3a.14
(20) (a) Bender, C. F.; Widenhoefer, R. A. Chem. Commun. 2008, 2741.
(b) Penzien, J.; Su, R. Q.; Mu¨ller, T. E. J. Mol. Catal., A 2002, 182-
183, 489. (c) Mu¨ller, T. E.; Berger, M.; Grosche, M.; Herdtweck, E.;
Schmidtchen, F. P. Organometallics 2001, 20, 4384.
(21) Kova´cs, G.; Ujaque, G.; Lledo´s, A. J. Am. Chem. Soc. 2008, 130,
853.
Intramolecular Hydroamination of Primary Amines. The
cyclohydroamination of unactivated alkenes with primary
amines provides an atom-economical synthetic route to N-H
containing cyclic secondary amines which, unlike alternative
protocols, circumvents the installation and removal of nitrogen
protecting groups. Although transformations of this type are well
established for several classes of hydroamination catalysts,1a
the late metal-mediated cyclization of such substrates has been
achieved only recently by use of [Rh(COD)2]BF4/biarylphos-
phine12 or Cu(OtBu)/Xantphos.15 Our initial efforts to effect the
cyclization of the primary aminoalkenes featured in Table 4 by
using [Ir(COD)Cl]2 as a precatalyst under a variety of experi-
mental conditions were unsuccessful. However, the knowledge
(22) (a) Sakai, N.; Ridder, A.; Hartwig, J. F. J. Am. Chem. Soc. 2006, 128,
8134. (b) Johns, A. M.; Utsunomiya, M.; Incarvito, C. D.; Hartwig,
J. F. J. Am. Chem. Soc. 2006, 128, 1828. (c) Pawlas, J.; Nakao, Y.;
Kawatsura, M.; Hartwig, J. F. J. Am. Chem. Soc. 2002, 124, 3669.
(d) Vo, L. K.; Singleton, D. A. Org. Lett. 2004, 6, 2469. (e)
Nettekoven, U.; Hartwig, J. F. J. Am. Chem. Soc. 2002, 124, 1166.
(23) Takaya, J.; Hartwig, J. F. J. Am. Chem. Soc. 2005, 127, 5756.
(24) McBee, J. L.; Bell, A. T.; Tilley, T. D. J. Am. Chem. Soc. 2008, 130,
16562.
(25) (a) Leighton-Munro, C.; Delp, S. A.; Alsop, N. M.; Blue, E. D.;
Gunnoe, T. B. Chem. Commun. 2008, 111. (b) Leighton-Munro, C.;
Delp, S. A.; Blue, E. D.; Gunnoe, T. B. Organometallics 2007, 26,
1483.
(26) Genin, E.; Antoniotti, S.; Michelet, V.; Geneˆt, J.-P. Angew. Chem.,
Int. Ed. 2005, 44, 4949.
9
J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010 417

A R T I C L E S
Hesp et al.
Figure 3. Relationship between kobs and 1a. The reactions were conducted
in 1,4-dioxane at 110 °C at the following concentrations: [Ir] ) 2.5 × 10-6
M and [1a] ) 2.0 × 10-4 M to 1.5 × 10-3 M.
Figure 1. Representative plot for the conversion of 1a to 2a catalyzed by
[Ir(COD)Cl]2. The reactions were conducted in 1,4-dioxane at 110 °C at
the following concentrations: [Ir] ) 5.0 × 10-6 M and [1a] ) 5.0 × 10-4
M.
Figure 4. Hammett plot for the [Ir(COD)Cl]2 catalyzed hydroamination
of arylamines 3a-e. The reactions were conducted in 1,4-dioxane at 110
°C at the following concentrations: [3a-e] ) 5.0 × 10-4 M and [Ir] ) 5.0
× 10-6 M.
Figure 2. Relationship between kobs and [Ir]. The reactions were conducted
in 1,4-dioxane at 110 °C at the following concentrations: [1a] ) 5.0 ×
10-4 M and [Ir] ) 1.0 × 10-6 M to 4.75 × 10-6 M.
unactivated alkenes with secondary alkyl- or arylamines has yet
to appear in the literature. In an effort to gain insight into the
mechanism of such transformations, so as to guide the develop-
ment of increasingly effective catalysts, a kinetic examination
of the [Ir(COD)Cl]2-mediated hydroamination reactions featured
herein was conducted.
to f(x) ) a[Ir]n, which provided an order (n) of 1.15(9) (Figure
S4b, Supporting Information).
The kinetic order in 1a was investigated subsequently by
using a constant [Ir] (0.0025 mM) along with a [1a] ranging
from 0.25 to 1.5 mM in 1,4-dioxane at 110 °C. In keeping with
the findings of Cochran and Michael,8a we observed decreasing
reaction rates with increased initial concentrations of 1a, which
is consistent with an inverse order dependence of the rate of
reaction on substrate concentration (Figure 3). Furthermore, the
rate of cyclization of 1a was observed to decrease by ap-
proximately 50% when measured in the presence of an
equivalent of added 2a. These results demonstrate both competi-
tive substrate and product inhibition. While on the basis of these
kinetic data alone we are unable to comment definitively
regarding the mode of catalyst inhibition by 1a and 2a, possible
scenarios that can be envisioned include (but are not restricted
to) unproductive binding of 1a or 2a to Ir, as well as
deprotonation of an amine/ammonium intermediate8b by 1a or
2a. However, computational studies support catalyst inhibition
by 1a and 2a arising from the latter deprotonation pathway (Vide
infra).
Kinetic Order of the Reaction. Kinetic parameters were
obtained for the hydroamination of 1a catalyzed by [Ir-
(COD)Cl]2; pseudofirst order rate constants were determined
by monitoring the disappearance and appearance of 1a and 2a,
respectively, relative to an internal standard over the course of
2-3 half-lives (1H NMR; Figure 1). We first examined the order
of reaction in [Ir(COD)Cl]2 by combining 1a (0.5 mM) with
varied concentrations of [Ir] (0.001 to 0.00475 mM) in 1,4-
dioxane at 110 °C. As shown in Figure 2, a plot of reaction
rate (kobs) versus [Ir] was linear, indicating a first-order kinetic
dependence on catalyst. This observation is consistent with a
mechanism involving a monomeric Ir catalyst species, and
would appear to preclude slow dissociation of a dimeric
precatalyst to an active monomer. Further confirmation of a first-
order dependence was provided by a nonlinear least-squares fit
9
418 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010

[Ir(COD)Cl]2 as a Catalyst Precursor
A R T I C L E S
Figure 5. (a) Eyring plot for the hydroamination of 1a catalyzed by [Ir(COD)Cl]2. The reactions were conducted in 1,4-dioxane over a temperature range
of 90-130 °C at the following concentrations: [1a] ) 5.0 × 10-4 M and [Ir] ) 2.5 × 10-6 M (∆Hq ) 20.9(3) kcal mol-1, ∆Sq ) -23.1(8) cal/K ·mol). (b)
Arrhenius plot for the hydroamination of 1a catalyzed by [Ir(COD)Cl]2. The reactions were conducted in 1,4-dioxane over a temperature range of 90-130
°C at the following concentrations: [1a] ) 5.0 × 10-4 M and [Ir] ) 2.5 × 10-6 M (Ea ) 21.6(3) kcal mol-1).
Hammett Study. The rate of the [Ir(COD)Cl]2-catalyzed
hydroamination of 3a-e was examined as a function of the
para-substituent (X) on the arylamine (eq 2). The cyclization
of 3a-e (Table 3, entries 3-1 to 3-5) in the presence of 0.5
mol % [Ir(COD)Cl]2 was monitored by use of 1H NMR methods
to obtain reaction rates. A Hammett plot of the resulting data
provided a large, negative F value of -2.4(2) (Figure 4), which
indicates that more electron-rich arylamines undergo hydroami-
nation more rapidly.
large negative ∆Sq value in the cyclization 1a mediated by
[Ir(COD)Cl]2 is consistent with a highly ordered transition
state.27
Summary of Kinetic Experiments. Kinetic analyses of the
[Ir(COD)Cl]2-catalyzed hydroamination of 1a have revealed that
the reaction rate shows first order dependence on the concentra-
tion of Ir and inverse order dependence with respect to both
substrate (1a) and product (2a) concentrations. While the
observed primary kinetic isotope effect (kH/kD ) 3.4(3)) and
the observation that increased electron density at nitrogen in 3
promotes the cyclization of these arylaminoalkene substrates
(F ) -2.4) are in keeping with an alkene activation mechanism
involving nucleophilic attack of a tethered amine on a metal-
coordinated alkene followed by a rate-limiting protonolysis step,
alternative mechanisms including those involving N-H oxida-
tive addition cannot be ruled out on the basis of these empirical
data alone. Given the lack of spectroscopic evidence that would
enable the definitive identification of catalytic intermediates in
the present study (Vide supra), and in an effort to gain further
insight into the elusive details of the mechanism of the
[Ir(COD)Cl]2-catalyzed cyclohydroamination of unactivated
aminoalkenes, this process was subjected to computational
examination.
Kinetic Isotope Effect. To provide some insight into the
turnover-limiting step of the mechanism, the H/D kinetic isotope
effect (KIE) for the [Ir(COD)Cl]2-catalyzed hydroamination 1a
was measured under catalytic conditions. The pseudofirst order
plots (Figure S7, Supporting Information) of the cyclization of
1a and 1a-d1 in 1,4-dioxane at 110 °C (eq 3) provided kobs
)
0.30(1) h-1 and kobs ) 0.089(2) h-1, respectively, which
translated to a primary KIE of 3.4(3). While these observations
implicate the hydrogen derived from the N-H of 1a in the rate-
determining step, several competing reaction pathways (e.g.,
rate-limiting N-H oxidative addition or proton transfer follow-
ing nucleophilic attack on a coordinated alkene) that account
for these observations can be envisioned.
Computational Study. The DFT method has been employed
as an established and predictive means to aid mechanistic
understanding; the particular value of such methods for unravel-
ing intricacies of hydroamination catalysis has been demon-
strated previously by several groups.28-31 Plausible mechanistic
scenarios employing [Ir(COD)Cl]2 as a precatalyst involving
either activation of the alkene linkage (CdC bond activation
Activation Parameters. To obtain activation parameters for
the [Ir(COD)Cl]2-catalyzed hydroamination of 1a, a series of
reactions were executed over a range of temperatures (90 -
130 °C); pseudofirst order rate constants were obtained from
linear plots of [1a]t versus time over this temperature range.
Eyring and Arrhenius analyses for the cyclization of 1a to 2a
afforded ∆Hq ) 20.9(3) kcal mol-1, ∆Sq ) -23.1(8) cal/K ·mol,
and Ea ) 21.6(3) kcal mol-1 (Figure 5). The observation of a
(27) Anslyn, E. V.; Dougherty, D. A. Modern Physical Organic Chemistry,
1st ed.; University Science Books: Sausalito, 2006; p 372.
(28) For a computational study of group 8 metal-assisted intermolecular
hydroamination, see: (a) Senn, H. M.; Blo¨chl, P. E.; Togni, A. J. Am.
Chem. Soc. 2000, 122, 4098. (b) Tsipis, C. A.; Kefaldis, C. E.
Organometallics 2006, 25, 1696. (c) Tsipis, C. A.; Kefaldis, C. E. J.
Organomet. Chem. 2007, 692, 5245.
(29) For a computational study of group 4 metal-assisted intermolecular
hydroamination, see: Straub, B. F.; Bergman, R. G. Angew. Chem.,
Int. Ed. 2001, 40, 4632.
9
J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010 419

A R T I C L E S
Hesp et al.
Scheme 1. Plausible Mechanistic Pathways for Late Transition Metal-Catalyzed Cyclohydroamination, Exemplified for [Ir(COD)Cl]2 M0 as
Catalyst Precursor and 5a as Substrate
route, Scheme 1, left)32 or of the amine functionality (N-H
bond activation route, Scheme 1, right)33 have been examined
for the cyclohydroamination of N-benzyl-2,2-dimethyl-4-en-1-
amine 5a (representing a prototypical secondary alkylamine)
into the pyrrolidine 6a (entry 2-7 in Table 2). Notably, this
represents the first attempt to comprehensively address the
mechanism of intramolecular hydroamination mediated by late
transition metal compounds by means of computational methods.
No structural simplification of any of the key species has been
imposed. While the presentation herein is focused exclusively
(30) For computational studies of group 4 transition metal-assisted in-
Figure 6. Various forms of the catalytically competent compound M1 (free
energy in kcal mol-1 relative to M1a).
tramolecular hydroamination of various substrate classes, see: ami-
noalkene substrates: (a) Mu¨ller, C.; Koch, R.; Doye, S. Chem.-Eur.
J. 2008, 14, 10430. aminoallene substrates: (b) Tobisch, S. Dalton
Trans. 2006, 4277. (c) Tobisch, S. Chem.sEur. J. 2007, 13, 4884.
(d) Tobisch, S. Chem.sEur. J. 2008, 14, 8590.
on the most accessible pathways, a complete account of all
scrutinized species is given in the Supporting Information.34
Nature of the Catalytically Competent Species. We start by
examining the fragmentation of precatalyst M0 in the presence
of substrate 5a and 1,4-dioxane solvent,35 which leads to the
catalytically competent species. DFT predicts that [Ir(COD)-
Cl(substrate)] complex M1 is predominantly generated in a
transformation that is slightly downhill (∆G ) -3.1 kcal mol-1
for 1/2M0 + 5a f M1a), while [Ir(COD)Cl(dioxane)] M1*
and dioxane or substrate adducts of M1 or M1* (respectively)
are distinctly disfavored thermodynamically.34 Compound M1
is present in several forms featuring a chelating aminoalkene
(M1a), or a monohapto association to the IrI center through its
amine (M1b) or olefin (M1c) functionalities, all of which are
in rapid equilibrium (Figure 6).36 Although M1a is most stable,
it does not represent the direct precursor for cyclohydroamina-
tion proceeding through rival routes for olefin or N-H bond
(31) For computational studies of organolanthanide-assisted intramolecular
hydroamination of various substrate classes, see: aminoalkene sub-
strates: (a) Motta, A.; Lanza, G.; Fragala, I. L. Organometallics 2004,
23, 4097. aminoalkyne substrates: (b) Motta, A.; Lanza, G.; Fragala,
I. L. Organometallics 2006, 25, 5533. aminoallene substrate: (c)
Tobisch, S. Chem.sEur. J. 2006, 12, 2520. aminodiene substrates: (d)
Tobisch, S. J. Am. Chem. Soc. 2005, 127, 11979. (e) Tobisch, S.
Chem.sEur. J. 2007, 13, 9127.
(32) For a selection of lead references for the olefin activation route, see:
(a) Kiji, J.; Nishimura, S.; Yoshikawa, S.; Sasakawa, E.; Furukawa,
J. Bull. Chem. Soc. Jpn. 1974, 47, 2523. (b) Armbu¨chl, J.; Pregosin,
P. S.; Venanzi, L. M.; Consiglio, G.; Bachechi, F.; Zambonelli, L. J.
Organomet. Chem. 1979, 181, 255. (c) Seligson, A. L.; Trogler, W. C.
Organometallics 1993, 12, 744. (d) Kwatsura, M.; Hartwig, J. F. J. Am.
Chem. Soc. 2000, 122, 9646. (e) Mu¨ller, T. E.; Grosche, M.;
Herdtweck, E.; Pleier, A.-K.; Walter, E.; Yan, Y.-K. Organometallics
2000, 1, 170. (f) Seul, J. M.; Park, S. J. Chem. Soc., Dalton Trans.
2002, 1153. (g) Karshtedt, D.; Bell, A. T.; Tilley, T. D. J. Am. Chem.
Soc. 2005, 127, 12640. (h) References 19e and 20c herein.
(33) For a selection of lead references for the amine activation route, see:
(a) Casalnuovo, A. L.; Calabrese, J. C.; Milstein, D. Inorg. Chem.
1987, 26, 971. (b) Cowan, R. L.; Trogler, W. C. Organometallics 1987,
6, 2451. (c) Park, S.; Johnson, M. P.; Roundhill, D. M. Organome-
tallics 1989, 8, 1700. (d) Ladipo, F. T.; Merola, J. S. Inorg. Chem.
1990, 29, 4127. (e) Koelliker, R.; Milstein, D. Angew. Chem., Int.
Ed. Engl. 1991, 30, 707. (f) Glueck, D. S.; Newman Winslow, L. J.;
Bergman, R. G. Organometallics 1991, 10, 1461. (g) Seligson, A. L.;
Cowan, R. L.; Trogler, W. C. Inorg. Chem. 1991, 30, 3371. (h)
References 16-18 herein.
(34) See the Supporting Information for more detail.
(35) The activation energy for fragmentation of M0 has not been assessed,
but it is rather unlikely to invoke any significant barrier, as the
examination by a linear-transit approach gave no indication for the
existence of such a barrier.
(36) Examination by a linear-transit approach gave no indication that this
process is associated with a substantial enthalpy barrier.
9
420 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010

[Ir(COD)Cl]2 as a Catalyst Precursor
A R T I C L E S
Figure 7. Free energy profile for the intramolecular hydroamination of 5a by [Ir(COD)Cl]2 (M0) proceeding via CdC bond activation. Only the most
accessible pathway for all principal steps is shown, while alternative, but less favorable pathways are omitted for the sake of clarity.
activation (Vide infra). Notably, the energetic cost of associating
a second equivalent of substrate as in M1b-S suggests that
substrate inhibition via formation of [Ir(COD)(substrate)2]Cl
species is unlikely.
gives rise to M2a, which has the hydrogen of the quarternized
ammonium center placed closely to the IrI center, while the
approach from backside leads first to M2b. Both transient
intermediates M2a and M2b relax almost instantaneously into
the minimum-energy conformer M2c,36 featuring a positively
charged ammonium unit that points toward the negatively
polarized chlorine. Intramolecular C-N bond formation does
preferably proceed via external attack of the amine unit on the
coordinated double bond and commences from M1c as the direct
precursor species, such that the amine group must first discon-
nect from the Ir center in the prevalent isomer M1a prior to
cyclization. Both trajectories are found to be viable, with
backside attack emerging as the dominant pathway in a process
that is uphill and features a moderate activation barrier (∆Gq )
16.9 kcal mol-1), thereby indicating that ring closure is smooth
and reversible.
Several scenarios for cleavage of the Ir-C bond in M2 have
been probed, including: (a) direct protonolysis by the proton
available from the ammonium unit; and (b) proton transfer to
the Ir center, affording the IrIII-hydrido-cycloamido intermediate
M3, with subsequent C-H reductive elimination of the cy-
cloamine.34 The stepwise mechanism via transient M3 represents
the kinetically advantageous scenario. The most accessible
pathway commences from M2a,38 which has the ammonium
unit already suitably oriented toward the Ir center, such that its
deprotonation does not cause major structural reorganization.
Moreover, the nucleophilicity of the IrI center supports this
transformation electronically. Indeed, the M2a f M3a proton
transfer has a low barrier of only 10.8 kcal mol-1 (relative to
the most stable M2c) to overcome and leads directly to the
square-pyramidal M3a bearing the hydride in the apical position.
The observed similarities in hydroamination of 1a mediated
by [Ir(COD)Cl]2 and [Ir(COD)Cl2]/2 AgBF4 (Vide supra) may
argue in favor of a formally cationic complex as representing
the active catalyst species. We have therefore probed the aptitude
of the dioxane solvent to support the displacement of the
chloride from the immediate proximity of the Ir center.37 Given
the low acceptor number of dioxane, it comes as no surprise
that a [Ir(COD)(substrate)(dioxane)]Cl solvent-separated ion pair
is found to be distinctly less stable than M1, being uphill by
more than 28 kcal mol-1,34 which corroborates the conductivity
measurements (Vide supra). Thus, our combined experimental
and computational approach provides compelling evidence for
M1 as being the catalytically competent compound.
Hydroamination via CdC Bond Activation. Principal steps
of the CdC bond activation route include: (1) nucleophilic attack
of the amine on the olefin functionality, which is activated by
coordination to the Ir center, to form a zwitterionic IrI-
cycloammonio-alkyl intermediate; (2) net protonolysis of the
Ir-C bond; and (3) displacement of cycloamine by a new
substrate molecule to regenerate M1. Figure 7 shows the Gibbs
free-energy profile, which considers only the most accessible
reaction pathways.34
The C-N bond forming step can proceed through distinct
trajectories constituting frontside (i.e., amine approach cis(syn)
to the metal) and backside (i.e., amine approach trans(anti) to
the metal) attacks, thereby giving rise to various conformers of
the zwitterionic IrI-cycloammonio-alkyl intermediate M2 (see
Figure 7). The cis(syn) amine attack on the olefin unit initially
(38) The various conformers of the zwitterionic intermediate M2 are in
rapid equilibrium, as mutual interconversion does not impose any
substantial barrier due to almost free rotation of the azacycle unit
around Ir-C and C-C bonds in M2.
(37) This was done by employing a discrete ensemble of three molecules
of 1,4-dioxane closely interacting with the Cl center, thereby simulating
specific solvation effects adequately.
9
J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010 421

A R T I C L E S
Hesp et al.
Scheme 2. ꢀ-Hydride Elimination as a Rival Event for the CdC
Bond Activation Route
may serve in rationalizing why the formation of pyrrolidines
other than 6a is not observed when using [Ir(COD)Cl]2 as a
precatalyst.
Hydroamination via N-H Bond Activation. An alternative
hydroamination pathway involving N-H bond activation (Scheme
1, right) features the following elementary steps: (1) oxidative
addition of the amine N-H linkage to the IrI center; (2) insertion
of the olefin unit into either Ir-N or Ir-H bonds of the resulting
IrIII-amido-hydrido complex; (3) generation of the cycloamine
by either C-H or C-N reductive elimination; and (4) liberation
of the cycloamine by new substrate to regenerate M1. The
condensed Gibbs free-energy profile, considering only the most
accessible pathways for relevant steps, is displayed in Figure
8.
Commencing from the active compound M1b, oxidative
addition of the amine N-H bond at the low-valent, coordina-
tively unsaturated IrI center initially gives rise to the five-
coordinate IrIII-η1-amidoalkene-hydrido compound M6a. The
positional isomer participating along the most accessible
pathway is depicted in Figure 8.34 The square-planar structure
of precursor M1b remains virtually undisturbed in the TS
structure, which in turn decays into square-pyramidal M6a,
featuring hydride in an axial position. Given the rather electron-
deficient nature of the IrI center in M1b, it comes as no surprise
that this oxidative addition step imposes an almost insurmount-
able barrier of 38.2 kcal mol-1 and is furthermore seen to be
strongly endergonic (∆G ) 25.7 kcal mol-1), thereby rendering
oxidative addition challenging on both kinetic and thermody-
namic grounds in the present system. The uncoordinated CdC
bond of the η1(N)-amidoalkene-Ir unit in M6a can approach
the metal center at its remaining coordination site, thereby giving
rise to the six-coordinate IrIII-η2-amidoalkene-hydrido intermedi-
ate M6′. However, this supposedly facile conversion36 comes
at some thermodynamic cost, as M6′ is less stable than M6.
The several possible isomers of M6′, two of which are shown
in Figure 8, are likely to interconvert readily.41
Two distinct paths for Ir-amidoalkene (M6′) f Ir-cycloamine
(M4b) transformation are conceivable, involving: (1) olefin
insertion into the IrIII-N bond of M6′ to afford IrIII-azacycle-
hydrido intermediate M7 with subsequent C-H reductive
elimination; or (2) insertion of the double bond into the IrIII-H
linkage of M6′ and ensuing C-N reductive elimination under
ring closure. The latter path is found to be more demanding
energetically and well separated from the former and remains
therefore virtually closed throughout the process.34 DFT unveils
that the dominant path toward pyrrolidine formation within this
mechanistic manifold entails ring closure by insertion of the
double bond into the Ir-N bond of M6′, followed by reductive
C-H elimination from M7. Notably, this pathway is consistent
with reports from Milstein,17a Togni17b and Hartwig18 involving
catalysts featuring a more electron-rich (bisphosphine)Ir frag-
ment, which are known to proceed via activation of the N-H
amine bond, and for which the barrier to N-H bond activation
is presumably lower than in M1b.
M3a represents a very shallow minimum on the free-energy
surface (with ∆Gq ) 0.4 kcal mol-1 for reverse M3a f M2a),
thereby implicating an almost negligible stationary population,
which would make M3a rather difficult to observe by use of
NMR spectroscopic techniques (Vide supra). Transient M3a can
undergo skeletal rearrangement39 to afford six-coordinate M3b
as the most stable form of the IrIII-hydrido complex, having Cl
and H in trans positions and the azacycle’s N donor center
directly attached to the metal (Figure 7). Notably, the transition
state (TS) structure traversed along the most accessible pathway
for reductive elimination adopts the same ligand arrangement.
Given the small energy gap between the initially formed M3a
and the lowest-energy TS for reductive elimination, it can
reasonably be assumed that skeletal rearrangement in M3a and
ensuing reductive elimination do occur concomitantly. Accord-
ingly, C-H bond reductive elimination to commence from five-
coordinate M3a proceeds through a highly ordered six-
coordinate TS structure and affords the IrI-pyrrolidine compound
M4b. The associated low barrier (∆Gq ) 4.7 kcal mol-1
)
together with the substantial thermodynamic force (∆G ) 17.0
kcal mol-1) is in keeping with M3a as being a high-energy
reactive intermediate. The reductive elimination has the higher
total barrier (i.e., relative to M1a) of the two consecutive steps
and thus discriminates the overall kinetics of the stepwise
cleavage of the Ir-C bond in M2. Thereafter, incoming substrate
displaces the pyrrolidine 6a from M4b and regenerates the M1c
active species, which re-enters the catalytic cycle. The exchange
of cycloamine by substrate is thermodynamically favorable (∆G
) -9.3 kcal mol-1) and kinetically viable (∆Gq ) 19.7 kcal
mol-1) when following an associative mechanism.
Having characterized all principal steps thus far, the precise
knowledge of other steps that may lead to unwanted products
is also of interest for the future development of improved
catalysts. To this end, we have examined a diverging route that
starts from M2 involving ꢀ-hydride elimination. The most
accessible pathway traverses a four-membered square-pyramidal
TS structure, featuring cis disposed olefin and hydride units,
which decays via smooth skeletal rearrangements into trigonal-
bipyramidal Ir-cycloenammonium-hydrido compound M5a
(Scheme 2).34 Compared to the rival pathway, namely stepwise
M2a f M3a f M4b Ir-C bond cleavage, the M2c f M5a
ꢀ-hydride elimination is predicted to be distinctly more demand-
ing kinetically and is therefore not accessible.40 These findings
Ring closure in M6′ through double bond insertion into the
IrIII-N bond has a modest barrier of 10.7 kcal mol-1 and is
strongly exergonic. The pathway that involves the thermody-
namically favorable M6’c, where H and Cl occupy trans
positions, is seen to be most accessible on both thermodynamic
and kinetic grounds. Notably, the directly formed IrIII-azacycle-
(39) We have not determined the barrier for skeletal (polytopal) rearrange-
ment between different forms of M3, but such processes are known
to be facile.
(40) DFT predicts a kinetic gap of 11.9 kcal mol-1 (∆∆Gq) between
stepwise protonolysis of the Ir-C bond and ꢀ-hydride elimination
commencing from M2, in favor of the former, which leads us to
confidently conclude that ꢀ-hydride elimination is not traversable.
(41) Note that the barriers for mutual interconversion via Berry rotation
are generally considered to be low.
9
422 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010

[Ir(COD)Cl]2 as a Catalyst Precursor
A R T I C L E S
Figure 8. Free energy profile for the N-H bond activation route of intramolecular hydroamination of 5a by [Ir(COD)Cl]2 (M0). Only the most accessible
pathway for all principal steps is shown, while alternative, but less favorable pathways are omitted for the sake of clarity.
Scheme 3. Proposed Mechanism for the [Ir(COD)Cl]2-Mediated
Intramolecular Hydroamination of Aminoalkenes
hydrido species M7c must not undergo skeletal rearrangements
prior to C-H reductive elimination. The lowest-energy pathway
commencing from M7c has a barrier of 18.5 kcal mol-1 to afford
the IrI-pyrrolidine compound M4b, which is downhill by another
9.5 kcal mol-1. Regeneration of M1b via associative displace-
ment of pyrrolidine by substrate from M4b is almost identical
energetically to the trajectory analyzed above that leads to M1c.
Hence, if the amine N-H bond were successfully to add to
the IrI center (as in (bisphosphine)Ir species16-18), subsequent
conversion of the amidoalkene unit into pyrrolidine via M6’c
f M7c f M4b would be kinetically viable and strongly
downhill. As revealed from Figure 8 the first oxidative addition
has the highest barrier and thus discriminates whether the N-H
bond activation route can be traversed. The assessed unfavorable
profile, featuring an almost insurmountable barrier of 38.3 kcal
mol-1 and being strongly uphill, is not compatible with the
smooth hydroamination reported herein, and mitigates against
N-H bond activation as being operative. This leads us to
conclude that the route via initial amine activation (Scheme 1,
right) is not accessible for the catalytically competent compound
M1 that features a rather electron-poor IrI center, and instead
the route via olefin activation (Scheme 1, left) prevails.
Mechanistic Conclusions. Our collective experimental and
computational results lead us to the following mechanistic
conclusions (Scheme 3). (1) Electrophilic activation of the CdC
linkage by a relatively electron-poor IrI center toward nucleo-
philic attack from the amine unit emerges as the operative
mechanism for the [Ir(COD)Cl]2-catalyzed cyclohydroamination
of unactivated aminoalkenes. (2) The [Ir(COD)Cl(substrate)]
complex M1 is the catalytically competent compound. (3) The
most stable form of [Ir(COD)Cl(substrate)] is M1a, which
features a chelating aminoalkene; M1a is the most abundant
intermediate and therefore likely represents the catalyst resting
state. Although M1a is a dormant species, it is easily converted
into M1c, featuring η1-olefin-Ir ligation, to start the productive
catalytic cycle. (4) Ring closure via nucleophilic attack of the
amine unit at the olefin functionality that is activated by
coordination to the Ir center has a moderate barrier and is
reversible. (5) Stepwise transfer of the proton from the am-
monium unit in M2a onto the Ir center with subsequent C-H
reductive elimination is the dominant pathway for cleavage of
the Ir-C bond in the zwitterionic intermediate. The proton
transfer step is kinetically viable and leads to the weakly
9
J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010 423

A R T I C L E S
Hesp et al.
stabilized, transient five-coordinate IrIII-hydrido intermediate
M3a, which thereafter undergoes reductive elimination by
traversing a highly ordered six-coordinate TS structure. (6) The
total activation energy for reductive elimination42a is highest
among all relevant steps and as such this step must thus be
considered as being turnover limiting. (7) Ensuing liberation
of the cycloamine by incoming substrate is viable when
following an associative pathway and regenerates the catalyti-
cally active M1c in an exergonic, hence irreversible step. In
the context of this mechanistic scenario, the catalyst inhibition
observed empirically for 1a, 2a, and DBU may be attributable
in part to the action of these species as Brønsted bases toward
the ammonium unit in M2. Alternatively, coordination of DBU
to Ir could prevent substrate association in a manner that cannot
be achieved by the weaker pyrrolidine base 2a,42b thereby giving
rise to a negligible population of the active compound M1.
While the proposed olefin activation mechanism outlined in
Scheme 3 is unprecedented for Ir-catalyzed alkene hydro-
amination,16-18 comprehensive synthetic and mechanistic studies
by Cochran and Michael8a support the operation of a related
olefin activation pathway in the cyclohydroamination of unac-
tivated alkenes bearing tethered amide/carbamate fragments
mediated by cationic (PNP)Pd complexes. In keeping with our
observations detailed herein, these workers noted a strong
inverse dependence of the reaction rate on substrate concentra-
computational investigation was conducted; notably, this rep-
resents the first such combined mechanistic study of the late
metal-mediated cyclohydroamination of unactivated alkenes with
secondary alkyl- and arylamines to appear in the literature.
Collectively, the results of these studies reveal that these
intramolecular hydroamination reactions commence from [IrI-
(COD)Cl(substrate)] as the catalytically competent compound
and proceed via initial electrophilic activation of the olefin unit
by a relatively electron-poor metal center, in a manner that has
previously not been documented for Ir-catalyzed alkene hy-
droamination. An alternative mechanism involving activation
of the amine functionality, which is firmly established for
electron-rich Ir complexes, has proven inaccessible for the
catalyst system under investigation herein. The operative
mechanistic scenario involves: (1) smooth, reversible ring
closure via nucleophilic attack of the amine on the metal
coordinated, thus activated CdC bond; (2) stepwise cleavage
of the Ir-C bond in the zwitterionic intermediate via proton
transfer from the ammonium unit to Ir, with subsequent
turnover-limiting C-H reductive elimination; and (3) associative
displacement of the cycloamine by new substrate, thereby
regenerating the active catalyst species. Turnover-limiting
reductive elimination to involve a metastable, highly reactive
IrIII-hydride intermediate and traversing a highly ordered TS
structure is consistent with the measured primary H/D KIE and
the observed negative activation entropy. DFT predicts an
effective barrier for intramolecular hydroamination of a proto-
typical secondary alkylaminoalkene that is in remarkable
agreement with observed activation parameters. Given the small
number of late metal catalysts that have proven capable of
promoting the intramolecular addition of primary and/or second-
ary alkyl- and arylamines to unactivated olefins, as well as the
paucity of mechanistic data pertaining to these transformations,
the catalytic and mechanistic findings reported herein represent
an important contribution toward the development of increas-
ingly effective late metal catalysts for use in mediating cyclo-
hydroamination reactions.
tion arising due to the reversible deprotonation of
a
[(PNP)Pd(alkylammonium)]2+ intermediate by substrate, as well
as rate-limiting Pd-C bond cleavage (protonolysis) involving
this intermediate.8a
The assessed energy profile (Figure 7) is consonant with our
observed empirical data. Turnover-limiting reductive elimination
is consistent with the located highly organized TS structure and
rationalizes the negative activation entropy measured. The lack
of spectroscopic evidence for the putative IrIII-hydrido species
M3a is understandable on the basis of its identification as a
metastable intermediate, and rate-limiting reductive elimination
from this intermediate is in keeping with the measured primary
H/D KIE. Productive ꢀ-hydrogen elimination, as a potentially
competitive event to productive hydroamination, proved not to
be viable on the basis of computational data, thereby supporting
the clean and efficient catalysis observed for several substrates.
Finally, the assessed total barrier42a of 24.6 kcal mol-1 for
turnover-limiting reductive elimination agrees remarkably well
with Eyring data for the cyclization of 1a to 2a (Vide supra).42c
Experimental Section
General Considerations. All manipulations were conducted in
the absence of oxygen and water under an atmosphere of dinitrogen,
either by use of standard Schlenk methods or within an mBraun
glovebox apparatus, utilizing glassware that was oven-dried (130
°C) and evacuated while hot prior to use. 1,4-Dioxane (Aldrich)
was dried over Na/benzophenone followed by distillation under an
atmosphere of dinitrogen. Toluene was deoxygenated by sparging
with dinitrogen followed by passage through a double column
solvent purification system purchased from mBraun Inc. 1,2-
Dimethoxyethane and 1,2-dichloroethane were deoxygenated by
sparging with dinitrogen gas followed by storage over activated 4
Å molecular sieves for 48 h prior to use. Chloroform-d1 (Cambridge
Isotopes) was used as received. All solvents used within the
glovebox were stored over activated 4 Å molecular sieves. With
Summary and Conclusions
We have identified [Ir(COD)Cl]2 as a highly effective
precatalyst for the intramolecular addition of primary as well
as secondary alkyl- or arylamines to tethered unactivated olefins
at relatively low catalyst loading. With the exception of the
hydroamination of primary aminoalkenes, for which HNEt3Cl
was required as a cocatalyst, no coligands or additives were
required in order to promote the hydroamination of these
challenging substrates. In the quest to develop mechanistic
insight into these intramolecular hydroamination reactions
catalyzed by [Ir(COD)Cl]2, a comprehensive experimental and
43
the exception of [Ir(COD)2]+BF4- and [Ir(MeCN)2(COD)]+BF4
,
-
as well as 2-(di-tert-butylphosphino)-N,N-dimethylaniline44 and
[Pd(cinnamyl)Cl]2,45 all of which were prepared using literature
procedures, all other Rh and Ir complexes were purchased from
Strem Chemicals and were evacuated under reduced pressure for
24 h prior to use. LiB(C6F5)4 ·2.5OEt2 (Boulder Scientific), LiOTf
(Strem), and AgBF4 (Strem) were evacuated under reduced pressure
(42) (a) The absolute TOF-determining barrier is given relative to the
catalyst resting state M1a. (b) DFT predicts that DBU binds more
strongly in [Ir(COD)Cl(amine)] compounds than 1a does in M1a by
12.0 kcal mol-1. On the other hand, the aminoalkene 1a displaces
pyrrolidine 2a from M4b in a kinetically viable, downhill step (see
Figure 7). (c) Cyclohydroamination of 1a to 2a has an observed
turnover-limiting barrier of 27.7 kcal mol-1 (∆Gq, 298 K).
(43) Green, M.; Kuc, T. A.; Taylor, S. H. Chem. Commun. 1970, 1553.
(44) Lundgren, R. J.; Stradiotto, M. Chem.sEur. J. 2008, 14, 10388.
(45) Auburn, P. R.; Mackenzie, P. B.; Bosnich, B. J. Am. Chem. Soc. 1985,
107, 2033.
9
424 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010

[Ir(COD)Cl]2 as a Catalyst Precursor
A R T I C L E S
for 24 h prior to use. 2,2-Diphenylpent-4-en-1-amine,11a (1-
allylcyclohexyl)methanamine,11a 2,2-diphenylhex-5-en-1-amine,12
4-methyl-2,2-diphenylpent-4-en-1-amine,11a N-benzyl-2,2-diphe-
nylpent-4-en-1-amine,11a N-(4-chlorobenzyl)-2,2-diphenylpent-4-
flash column chromatography on silica gel (hexanes/EtOAc ) 10:
1) to afford 1-phenyl-2-methyl-4,4-diphenylpyrrolidine as a color-
less oil that solidified to a white solid upon standing (74 mg, 0.22
mmol, 95%), and for which obtained characterization data agreed
with data reported in the literature.14
en-1-amine,6 a
methyl
4-[(2,2-diphenylpent-4-
enylamino)methyl]benzoate,11aN-(4-methoxybenzyl)-2,2-diphenylpent-
4-en-1-amine,6a N-(cyclohexylmethyl)-2,2-diphenylpent-4-en-1-
amine,12 (1-allylcyclohexyl)-N-benzylmethanamine,11a N-benzyl-
2,2-dimethylpent-4-en-1-amine,11a 2-phenyl-2-(2-propenyl)-1-amino-
4-pentene,46 N-benzyl-2-isopropylpent-4-en-1-amine,11a N-benzyl-
2,2-diphenylhex-5-en-1-amine,12 N-benzyl-4-methyl-2,2-diphenyl-
Representative Procedure for the Intramolecular Hydroami-
nation of Unactivated Alkenes by Primary Amines (Table 4).
To a screw-capped vial containing 2,2-diphenylpent-4-en-1-amine
(59 mg, 0.25 mmol), triethylammonium chloride (1.7 mg, 0.0125
mmol) and a stir-bar was added 0.5 mL of a stock solution of
[Ir(COD)Cl]2 (21 mg, 0.031 mmol) in 2.548 mL of 1,4-dioxane
([Ir] ) 0.025 mM). The vial was sealed under N2 with a cap
containing a PTFE septum and, once all the material had dissolved,
was removed from the glovebox and was placed in a temperature-
controlled aluminum heating block set at 110 °C. After 24 h of
magnetic stirring the vial was removed from the temperature-
controlled aluminum heating block, cooled to ambient temperature,
diluted with CH2Cl2 (2 mL), and was washed with brine (2 × 5
mL). The organic extracts were combined, dried over Na2SO4, and
concentrated. The resulting residue was purified by flash column
chromatography on silica gel (CH2Cl2/MeOH ) 10:1) to yield
2-methyl-4,4-diphenylpyrrolidine as a pale-yellow oil (53 mg, 0.22
mmol, 89%) that afforded analytical data in agreement with data
reported in the literature.12
Computational Section. All calculations based on Kohn-Sham
density functional theory (DFT)49 were performed by means of the
program package TURBOMOLE50 using the almost nonempirical
meta-GGA Tao-Perdue-Staroverov-Scuseria (TPSS) functional51
within the RI-J integral approximation52 in conjunction with flexible
basis sets of triple-ꢁ quality. For iridium we used the Stuttgart-
Dresden scalar-relativistic effective core potential (SDD, 60 core
electrons)53 in combination with the (8s7p6d1f)/[6s5p3d1f] (def2-
TZVP) valence basis set.54 All remaining elements were represented
by Ahlrich’s valence triple-ꢁ TZVP basis set55 with polarization
functions on all atoms. The good to excellent performance of the
TPSS functional for a wide range of applications, with transition-
metal complexes in particular, has been demonstrated previously.56
The energy landscape of the entire cyclohydroamination course was
assessed for the secondary alkylamine 5a together with [Ir-
(COD)Cl]2 M0 as precatalyst (entry 2-7 in Table 2). No structural
simplification of any of the involved key species was imposed. The
DFT calculations have simulated the authentic reaction conditions
by treating bulk effects of the 1,4-dioxane solvent by a consistent
polarizable continuum model.57 All the stationary points were fully
pent-4-en-1-amine, 4 7
benzyl[1-(2-methylallyl)-
cyclohexylmethyl]amine,11a (E)-N-benzyl-2,2-diphenylhex-4-en-1-
amine and (Z)-N-benzyl-2,2-diphenylhex-4-en-1-amine,14 tert-butyl-
2,2-diphenylpent-4-enylcarbamate,8c N-phenyl-2,2-diphenylpent-4-
en-1-amine,14 N-(4-methylphenyl)-2,2-diphenylpent-4-en-1-amine,14
N-(4-methoxyphenyl)-2,2-diphenylpent-4-en-1-amine,14 N-(4-fluo-
rophenyl)-2,2-diphenylpent-4-en-1-amine,14 (1-allylcyclohexyl)-N-
phenylmethanamine,48 N-phenyl-2,2-diphenylhex-5-en-1-amine,48
and N-phenyl-4-methyl-2,2-diphenylpent-4-en-1-amine,48 were syn-
thesized according to literature procedures. All other reagents were
used as received. 1H and 13C NMR characterization data were
collected at 300 K on a Bruker AV-500 spectrometer operating at
500.1 and 125.8 MHz (respectively) with chemical shifts reported
in parts per million downfield of SiMe4. Conductivity measurements
were carried out on a VWR SympHony pH/Conductivity meter
employing a VWR SympHony 4 carbon probe conductivity cell
with 0.55 cm-1 cell constant; calibrations were made with 1.12
microSeimens and 100,039 microSeimens Traceable conductivity
solutions. The curve fittings were carried out using the program
SigmaPlot for Windows v. 10.0 (Systat Software, San Jose, CA).
Representative Procedure for the Intramolecular Hydroami-
nation of Unactivated Alkenes by Secondary Alkylamines
(Table 2). To a screw-capped vial containing 1a (82 mg, 0.25
mmol) and a stir-bar was added 0.125 mL of a stock solution of
[Ir(COD)Cl]2 (2.6 mg, 0.0039 mmol) in 1.548 mL of 1,4-dioxane
([Ir] ) 0.005 mM) and 0.375 mL of dioxane (total reaction volume
) 0.5 mL). The vial was sealed under N2 with a cap containing a
PTFE septum and, once all the material had dissolved, was removed
from the glovebox and was placed in a temperature-controlled
aluminum heating block set at 110 °C. After 3 h of magnetic stirring
the vial was removed from the temperature-controlled aluminum
heating block, cooled to ambient temperature, diluted with CH2Cl2
(2 mL), and was washed with brine (2 × 5 mL). The organic
extracts were combined, dried over Na2SO4, and concentrated. The
resulting residue was purified by flash column chromatography on
silica gel (hexanes/EtOAc ) 20:1) to yield 2a as a white solid (72
mg, 0.22 mmol, 88%) that afforded analytical data in agreement
with data reported in the literature.11a
Representative Procedure for the Intramolecular Hydroami-
nation of Unactivated Alkenes by Arylamines (Table 3). To a
screw-capped vial containing N-phenyl-2,2-diphenylpent-4-en-1-
amine (78 mg, 0.25 mmol) and a stir-bar was added 0.5 mL of a
stock solution of [Ir(COD)Cl]2 (5.7 mg, 0.0085 mmol) in 1.357
mL of 1,4-dioxane ([Ir] ) 0.0125 mM). The vial was sealed under
N2 with a cap containing a PTFE septum and, once all the material
had dissolved, was removed from the glovebox and was placed in
a temperature-controlled aluminum heating block set at 110 °C.
After 7 h of magnetic stirring the vial was removed from the
temperature-controlled aluminum heating block, cooled to ambient
temperature, diluted with CH2Cl2 (2 mL), and was washed with
brine (2 × 5 mL). The organic extracts were combined, dried over
Na2SO4, and concentrated. The resulting residue was purified by
(49) Parr, R. G.; Yang, W. Density-Functional Theory of Atoms and
Molecules; Oxford University Press: Oxford, 1989.
(50) (a) Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.; Horn, H.; Ko¨lmel, C. Chem.
Phys. Lett. 1989, 162, 165. (b) Treutler, O.; Ahlrichs, R. J. Chem.
Phys. 1995, 102, 346. (c) Ahlrichs, R.; Furche, F.; Ha¨ttig, C.; Klopper,
W.; Sierka, M.; Weigend, F. TURBOMOLE, Version 6.0; University
of Karlsruhe: Karlsruhe, Germany, 2009; http://www.turbomole.com.
(51) (a) Dirac, P. A. M. Proc. R. Soc. London 1929, A123, 714. (b) Slater,
J. C. Phys. ReV. 1951, 81, 385. (c) Perdew, J. P.; Wang, Y. Phys.
ReV. 1992, B45, 13244. (d) Tao, J.; Perdew, J. P.; Staroverov, V. N.;
Scuseria, G. E. Phys. ReV. Lett. 2003, 91, 146401. (e) Perdew, J. P.;
Tao, J.; Staroverov, V. N.; Scuseria, G. E. J. Chem. Phys. 2004, 120,
6898.
(52) (a) Vahtras, O.; Almlo¨f, J.; Feyereisen, M. W. Chem. Phys. Lett. 1993,
¨
213, 514. (b) Eichkorn, K.; Treutler, O.; Ohm, H.; Ha¨ser, M.; Ahlrichs,
R. Chem. Phys. Lett. 1995, 242, 652.
(53) Andrae, D.; Ha¨ussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor.
Chim. Acta 1990, 77, 123.
(54) (a) Weigend, F.; Ahlrichs, R. Phys. Chem. Chem. Phys. 2005, 7, 3297.
(b) Weigend, F. Phys. Chem. Chem. Phys. 2006, 8, 1057.
(55) (a) Scha¨fer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
5829. (b) Eichkorn, K.; Weigend, F.; Treutler, O.; Ahlrichs, R. Theor.
Chem. Acc. 1997, 97, 119.
(46) Majumder, S.; Odom, A. L. Organometallics 2008, 27, 1174.
(47) Dochnahl, M.; Pissarek, J.-W.; Blechert, S.; Lo¨hnwitz, K.; Roesky,
P. W. Chem. Commun. 2006, 3405.
(48) Lundgren, R. J.; Sappong-Kumankumah, A.; Stradiotto, M.
Chem.sEur. J.
(56) (a) Staroverov, V. N.; Scuseria, G. E.; Tao, J.; Perdew, J. P. J. Chem.
Phys. 2003, 119, 12129. (b) Zao, Y.; Truhlar, D. G. J. Chem. Theory
Comput. 2005, 1, 415. (c) Furche, F.; Perdew, J. P. J. Chem. Phys.
2006, 124, 044103.
(57) Klamt, A.; Schu¨u¨rmann, G. J. Chem. Soc., Perkin Trans 2 1993, 799.
9
J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010 425

A R T I C L E S
Hesp et al.
located with inclusion of solvation. Analytical frequency calcula-
tions were performed to confirm the nature of all optimized key
structures and to determine thermodynamic parameters (298 K, 1
atm) under the rigid-rotor and harmonic approximations. The
mechanistic conclusions drawn in this study were based on the
Gibbs free-energy profile of the entire catalytic cycle assessed at
the TPSS(COSMO)/SDD+TZVP level of approximation for ex-
perimental condensed phase conditions. Further details together with
a description of the employed computational methodology are given
in the Supporting Information. Calculated structures were visualized
by employing the StrukEd program,58 which was also used for the
preparation of 3D molecule drawings.
for M. S. and a Canada Graduate Scholarship for K.D.H.) and
Dalhousie University (M.S. and K.D.H.) for their generous support
of this work. We also thank Dr. Michael Lumsden (NMR3,
Dalhousie) for assistance in the acquisition of NMR data, and Prof.
Heather Andreas (Dalhousie) for assistance in conducting solution
conductivity measurements.
Supporting Information Available: Complete experimental
procedures and characterization data, including tabulated ex-
perimental results and supporting NMR data, computational
details, detailed computational characterization of relevant
elementary steps, and Cartesian coordinates of all calculated
species. This material is available free of charge via the Internet
at http://pubs.acs.org.
Acknowledgment. We thank the Natural Sciences and Engi-
neering Research Council of Canada (including a Discovery Grant
(58) For further details, see http://www.struked.de.
JA908316N
9
426 J. AM. CHEM. SOC. VOL. 132, NO. 1, 2010