Optically active, bulky tris(oxazolinyl)borato magnesium and calcium compounds for asymmetric hydroamination/cyclization

Article abstract of DOI:10.1016/j.jorganchem.2010.08.057

The synthesis of the new chiral, pseudo C₃-symmetric, monoanionic ligand tris(4S-tert-butyl-2-oxazolinyl)phenylborate [To ^T]^- is reported. The steric bulk, tridentate coordination, and anionic charge of [To^T]

Full text of DOI:10.1016/j.jorganchem.2010.08.057

Journal of Organometallic Chemistry 696 (2011) 228e234
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Optically active, bulky tris(oxazolinyl)borato magnesium and calcium compounds
for asymmetric hydroamination/cyclization
Steven R. Neal, Arkady Ellern, Aaron D. Sadow*
Department of Chemistry, Iowa State University, Ames, IA 50011, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 31 July 2010
Received in revised form
The synthesis of the new chiral, pseudo C -symmetric, monoanionic ligand tris(4S-tert-butyl-2-oxa-
3
T ꢀ
zolinyl)phenylborate [To ] is reported. The steric bulk, tridentate coordination, and anionic charge of
T ꢀ
T
[
To ] are suitable for formation of complexes of the type To MX, where one valence is available for
30 August 2010
T
reactivity. With this point in mind, we prepared magnesium and calcium To complexes that resist
redistribution to (To ) M compounds. Both To MgMe and To CaC(SiHMe ) contain tridentate To -
2 2 3
Accepted 31 August 2010
Available online 17 September 2010
T
T
T
T
coordination to the metal center, as shown by NMR spectroscopy, infrared spectroscopy, and X-ray
crystallography. These compounds are active catalysts for the cyclization of three aminoalkenes to
pyrrolidines, and provide non-racemic mixtures of pyrrolidines in enantiomeric excesses up to 36%.
Ó 2010 Elsevier B.V. All rights reserved.
Keywords:
Hydroamination
Oxazoline
Alkaline earth metal
1. Introduction
complexes [5]. Additionally, interesting zwitterionic bis(carbene)-
and tris(carbene)borato alkaline earth metal compounds have been
Group 2 organometallic compounds have potential advantages in
homogeneous catalysis, as organomagnesium and organocalcium
compounds are inexpensive, their starting materials are readily
available, both metals are physiologically benign, and techniques for
their manipulation are typically similar to those developed for
Grignard reagents. However, their ionic bonding, facile configura-
tional and structural exchange reactions, and thermodynamic
stability of oxide and halide salts due to high lattice energies are
particular challenges that inhibit the application of Group 2 metal
compounds incatalysis. Inthisregard, thediscovery by Parkin and co-
workers that tris(pyrazolyl)boratomagnesium(II) compounds are
resistant toward disproportionation reactions and provide
compounds with well-defined coordination constitutions and
geometries offersopportunities for organomagnesium compounds in
reportedand applied inhydroamination/cyclization [6]. A few reports
of stereoselective hydroamination with calcium and magnesium-
based catalysts have provided % ee’s upto 6%and 14%, respectively [7].
We previously reported the synthesis of the achiral monoanionic
M ꢀ
tris(4,4-dimethyl-2-oxazolinyl)phenylborate [To ] and its chem-
istry in zirconium, [8] iridium, aluminum, and yttrium complexes,
[9] where the latter yttrium species are catalysts for the cyclization/
hydroamination of aminoalkenes to pyrrolidines [10].
These tridentate monoanionic ligands are electronically similar to
the well-known tris(pyrazolyl)borate ligands [11,12]. The tridentate
monoanionic oxazolinylborate ligands also may be compared to
Gade’s neutral tridentate tris(oxazolinyl)ethane ligands (trisox) [13]
whereas the anionic borate center in the tris(oxazolinyl)borates
provides an additional electrostatic component to their interaction
2þ
catalysis [1]. Later, Chisholm reported achiral C3v and chiral C
3
-
with metal centers. Compounds with the formula [{trisox}LnR]
(Ln ¼ Sc, Y, Lu, Tm, Er, Ho, Dy), generated in situ from {trisox}LnR
and [Ph C][B(C ], are catalysts for stereospecific polymerization
of -olefins [14]. Neutral tris(oxazolinyl)borate compounds of diva-
symmetric tris(pyrazolyl)boratomagnesium catalysts for lactide ring-
opening polymerization [2]. More recently, magnesium and calcium
diketiminate compounds (e.g. {nacnac}MR) have been shown to be
highly active in hydroamination/cyclization of aminoalkenes [3,4]
where sterically demanding substituents on the diketiminate
hinder the formation of bis(diketiminate)magnesium and calcium
3
3
6 5 4
F )
a
lent metal centers, such as magnesium, calcium, and zinc, are
isoelectronic with the putative dicationic trisox rare earth alkyls and
might provide reactive complexes. In this context, we recently
described a series of achiral tris(oxazolinyl)borato zinc compounds
containing chloride, hydride, alkoxide, and disilazide ligands [15].
We were curious if tris(oxazolinyl)borato magnesium and calcium
compounds might behave similarly, providing robust compounds
*
Corresponding author. Tel.: þ1 515 294 8069; fax: þ1 515 294 0105.
E-mail address: sadow@iastate.edu (A.D. Sadow).
0
022-328X/$ e see front matter Ó 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2010.08.057

S.R. Neal et al. / Journal of Organometallic Chemistry 696 (2011) 228e234
229
that could still access open coordination sites for chemical reactivity.
2D-OxtBu, and the 1H NMR spectrum of this mixture does not
contain resonances from diethyl ether or THF that might rationalize
Here we report the synthesis of tris(4S-tert-butyl-2-oxazolinyl)phe-
T ꢀ
tBu
1
nylborate [To ] , its complexes with magnesium(II) and calcium(II),
the high solubility of LiOx
in hydrocarbon solvents. A H NMR
tBu
and their reactivity as catalysts for the hydroamination/cyclization of
aminoalkenes.
spectrum of isolated LiOx in benzene-d
nances due to residual THF, and addition of THF-d
solution does not change the chemical shifts of the oxazolide in the
6
does not exhibit reso-
8
to this benzene
1
7
tBu
2
. Results and discussion
H NMR spectrum. Furthermore, the Li NMR spectrum of LiOx
(
1.22 ppm) in benzene-d
6
does not shift upon addition of THF.
tBu
2.1. Ligand synthesis
The IR spectrum of LiOx (KBr) contained
n
CN bands due to both
ꢀ1
ꢀ1
isocyanide (2000 cm ) and oxazolide (1635 cm ), and this is
consistent with previous observations with 4,4-dimethyl-2-lithio-
oxazolide and 4S-isopropyl-2-lithio-oxazolide [16,19].
M
The preparation of Li[To ] involves in situ deprotonation of 2H-
4
,4-dimethyl-2-oxazoline with nBuLi followed by reaction with
0
2
.30 equivalents of PhBCl (Eq. (1)) [8].
This high solubility contrasts 2Li-4,4-dimethyl-oxazolide
Me2
iPr
(
LiOx
uble in hydrocarbons, diethyl ether, and THF once isolated, though
the latter species dissolves in THF upon addition of PhBCl . The
isolated anion LiOx is insoluble in THF after isolation and was
) and 2Li-4S-isopropyl-oxazolide (LiOx ) that are insol-
We also prepared a chiral analog from the readily-available,
2
Me2
L
-valine-derived 4S-isopropyl-2-oxazoline to provide tris(4S-iso-
P ꢀ
M
propyl-2-oxazolinyl)phenylborate [To ] [16]. In contrast to the
synthesis of [To ] , deprotonation of 2H-4S-isopropyl-2-oxazoline
not useful for synthesis of Li[To ], although it may be used in situ in
THF. Thus, the procedure used for isolation of LiOx , which
M ꢀ
iPr
with nBuLi proceeds in lower yield (ca. 70%) [17] and treatment of the
3 2
involves washing with diethyl ether to remove the HN(SiMe ) , is
tBu
in situ prepared 2Li-4S-isopropyl-oxazolide with 0.31 equivalents of
not possible with LiOx . Neither repeated crystallization nor
P
ꢀ5
PhBCl
[
2
affords Li[To ] only irreproducibly. Instead, preparation of Li
exposure to high vacuum (10 mTorr) for extended times (2 days)
P
improved the purity of LiOxtBu.
To ] requires deprotonation of 2H-4S-isopropyl-2-oxazoline with
the bulky disilazide base LiN(SiMe , isolation of the 2Li-4S-iso-
propyl-oxazolide, and reaction of the isolated material with
.31 equivalents. of PhBCl (Eq. (2)). Thus, the seemingly straightfor-
)
3 2
Fortunately, a one-pot procedure, in which 2H-4S-tert-butyl-2-
ꢁ
oxazoline and tert-butyllithium are allowed to react inTHFat ꢀ78 C
ꢁ
0
2
followed by treatment with 0.31 equivalents of PhBCl
2
at ꢀ78
C
ward synthesis of tris(oxazolinyl)borates as proligands for organo-
metallic compounds has required optimization for each ligand, unlike
the corresponding bis(oxazolinyl)borates described by Pfaltz that are
affords the desired lithium tris(4S-tert-butyl-2-oxazolinyl)phenyl-
T
borate (Li[To ], Eq. (4)) as indicated by a single resonance at
d
ꢀ17.0
. The H NMR spectrum of
crude Li[To ] is consistent with formation of a C -symmetric product
as evident by one set of oxazoline resonances at
11
1
in the B NMR spectrum in methanol-d
4
T
prepared by in situ reaction of 2H-oxazolines, tBuLi, and Ph
2
BCl [18].
3
t
d
0.86 ( Bu), 3.75
tBu
(
CH), and 3.89 (CH ). Clearly, the yield of LiOx obtained from an in
2
situ 2H-4S-tert-butyl-2-oxazoline deprotonation with tert-butyl-
iPr
lithium is much improved in comparison to LiOx . Apparently, the
formation of a quaternary borate with three boron-oxazoline bonds
requires pure oxazolide anion and cannot tolerate the lower yield of
iPr
3 2
LiOx or small amounts of HN(SiMe ) .
T
A crude sample of Li[To ] (contaminated with LiCl) is converted
to the protonated hydrogen tris(4S-tert-butyl-2-oxazolinyl)phe-
T
nylborate H[To ] (Eq. (4)) by treatment with triethylammonium
Because development of new stereoselective catalytic reactions
often requires several structurally and electronically similar ligands
with a range of steric properties, we set out to prepare a tris(oxazo-
chloride in methylene chloride, followed by filtration through
alumina. Pure H[To ] is isolated after silica gel column chromatog-
T
raphy with the solvent mixture hexane:ethyl acetate:triethyl
t
linyl)borate derived from
L-tert-leucine, as the tert-butyl group will
amine ¼ 20:6:1. Only three oxazoline resonances
d
0.80 ( Bu), 3.44
provide an increased steric demand. The tridentate, monoanionic
nature of these tris(oxazolinyl)borate ligands appears well suited to
Group 2 chemistry because the ligandemetal interaction is sup-
ported by sterics and electrostatics, as well as coordination bonding.
3 2
Reaction of 2H-4S-tert-butyl-2-oxazoline and LiN(SiMe )
tBu
provides 2Li-4S-tert-butyl-2-oxazolide (LiOx ) (Eq. (3)). Although
tBu
formation of LiOx is quantitative in micromolar scale reactions,
3 2
its isolation from the HN(SiMe ) byproduct proved difficult due to
the high solubility of the oxazolide anion in pentane and diethyl
ether. No evidence could be obtained for THF coordination to
tBu
LiOx that might be responsible for the enhanced solubility of the
tert-butyl-oxazolide versus the 4S-isopropyl- or 4,4-dimethyl-
tBu
4
oxazolides. Reaction of LiOx with methanol-d as solvent forms

230
S.R. Neal et al. / Journal of Organometallic Chemistry 696 (2011) 228e234
(
(
CH), 3.78 (CH
2
) are observed in the 1H NMR spectrum of H[ToT]
benzene-d
6
3
), providing support for an optically pure, C -symmetric
P ꢀ
T
species. As in [To ] chemistry, the oxazoline groups in S,S,S-H[To ]
are optically pure because small amounts of 4R-tert-butyl-oxazoline
would give diastereomers (e.g. S,S,R) that would have a H NMR
1
T
spectrum that is distinct from the S,S,S-H[To ] diastereomer.
T
T
2
.2. Synthesis and characterization of To MgMe and To CaC
(
2 3
SiHMe )
N3
H[ToT] and MgMe
$(O C H ) react in benzene to afford
2 2 4 8 2
To MgMe (Eq. (5)). The oxazolines in To MgMe are equivalent in the
H NMR spectrum, indicating that the magnesium compound’s
B1
T
T
1
structure in solution is pseudo C
3
-symmetric.
Mg1
N1
N2
C26
Thus, one resonance, observed at 0.72 ppm, was assigned to
the tert-butyl moiety, and three resonances due to the
oxazoline methine and methylene were detected at 3.45, 3.59 and
Fig. 1. An ORTEP diagram of To
the stereogenic centers are shown to highlight the configuration, and the C
represented by a ball and stick structure (without the hydrogens) for clarity.
T
MgMe, drawn at 50% probability. Hydrogen atoms on
6 5
H
group is
3
.72 ppm. Additionally, the resonance due to the MgMe appeared
t
at ꢀ0.65 ppm; the integrated ratio of this resonance versus the Bu
2.137(7), and 2.137(7) Å are also essentially identical with those in
15
1
T
resonance was the expected 3 H:27 H. The N{ H} NMR chemical
To MgMe.[1] The NeMgeN and NeMgeC bond angles are similar
1
15
ꢁ
ꢁ
shift of ꢀ178.2 ppm was obtained from a He N HMBC experi-
ment, which contained a crosspeak between the oxazoline nitrogen
and oxazoline methine and methylene resonances. Unfortunately,
through-bond coupling (3 bonds) between the coordinated oxa-
(90.6(4)ꢀ91.3(3) ) and (122.7(4)e125.4(2) ) respectively. The
T tBu
similarity of MgeN bond lengths in To MgMe and Tp MgMe,
where 4S-tert-butyl-oxazolinyl and 3-tert-butyl-pyrazolyl are
expected to have distinct electronic properties, further emphasizes
the substantial ionic character of these four-coordinate magnesium
15
zoline nitrogen and magnesium methyl was not detected. The
N
1
{
H} NMR chemical shift is only slightly different than the value of
compounds.
T
better comparison between To MgMe and TptBuMgMe
T
ꢀ
174.5 ppm observed for H[To ], but it is significantly upfield from
A
tBu
that of 2H-4S-tert-butyl-2-oxazoline (ꢀ148.0 ppm). Evidence for
involves their relative steric properties. The cone angle of Tp was
previously estimated as 244 , [1] whereas the crystallographic
coordinates of To from To MgMe provide an estimation of its cone
angle as 233 . The cone angle measures the extent to which
T ꢀ
ꢁ
a three-point interaction between [To ] and magnesium (versus
T
T
rapid exchange of oxazoline in a bidentate ground state) is provided
ꢀ
T
1
ꢁ
by the infrared spectrum that contains only one
contrast, the
n
CN at 1585 cm . In
n
CN of 2H-4S-tert-butyl-2-oxazoline and H[To ] are
substituents on the pyrazole or oxazoline rings (i.e., tert-butyl)
extend past the magnesium center. A second approach for assessing
the steric properties of a ligand is provided by its solid angle, in
which the metal center is treated as a point source of light and the
complex is encompassed in an imaginary sphere [20]. The solid
angle is the surface area of a shadow cast by a ligand on the inside of
the sphere, providing an estimate of the overall steric impact of
ꢀ1
1
635 and 1601 cm , respectively [19] The steric impact of coor-
dination of the large [To ] ligand to the small magnesium(II) center
T
is sufficient to inhibit coordination of THF and dioxane.
T
X-ray quality crystals of To MgMe were obtained from
ꢁ
a concentrated toluene solution cooled to ꢀ78 C. A single crystal
diffraction study revealed that the tris(oxazolinyl)borate is coor-
dinated to the magnesium center in a tridentate fashion in the solid
T
tBu
a ligand. We have calculated the solid angles for To and Tp from
X-ray coordinates as 7.8 steradians (the percentage of the surface
area of the surrounding sphere ‘shaded’ is 62%) and 8.9 steradians
(71%), respectively, using the program Solid-G [21]. Taken together,
the cone angles and solid angles show that the effective size of the
3 3
state (Fig. 1). Pseudo-C symmetry is evident, where the C axis
coincides with the BeMg vector with the simplifying assumption
that the BeCphenyl is freely rotating in solution. Thus, the [To ]
ligand forms a propeller-type shape around the magnesium center.
T ꢀ
T
tBu
The chiral P2
only one enantiomer of To MgMe in the solid state.
1
2
1
2
1
space group is consistent with the presence of
To ligand is smaller than that of Tp . This steric difference is
related to the hybridization of the carbons on the heterocyclic rings,
T
2
The Mg1eC26 bond length is 2.102(1) Å; the Mg1eN1 and
Mg1eN3 bond lengths are equivalent (2.108(1) and 2.109(1) Å),
whereas the Mg1eN2 bond is slightly longer (2.118(1) Å). The steric
such that planar, sp -hybridized C3 in the pyrazole directs the tert-
butyl group past the magnesium center whereas sp hybridization
3
in the oxazoline ring points the tert-butyl group toward the back-
side of one of the adjacent oxazolines rather than in front of the
metal center. For example, the closest hydrogen on one tert-butyl
group is 2.5 Å from the nearest hydrogen on the back-side of an
adjacent oxazoline ring in the (static) conformation observed in the
solid state structure. Still, for a stereoselective reaction (such as an
insertion) in which a substrate must penetrate the space between
T
bulk and chelating [To ] ligand distort the bond angles of the Mg
center from tetrahedral, as expected for a highly ionic compound.
ꢁ
For example, the values for the :NeMgeN angles are close to 90
ꢁ
(
89.74(4)e91.56(4) ) whereas the :NeMgeC angles are more
ꢁ
open, ranging from 121.13(6) to 127.84(6) . For comparison, the
MgeC bond distance in the related tris(pyrazolyl)borate compound
tBu
tBu
T
Tp MgMe (Tp ¼ tris(3-tert-butyl-pyrazolyl)borate) of 2.12(1) Å
two oxazoline rings, the steric properties of To are expected to
is the same within 3 error, and the MgeN bond distances 2.13(1),
s
make a significant steric distinction between prochiral faces.

S.R. Neal et al. / Journal of Organometallic Chemistry 696 (2011) 228e234
231
T
tBu
Furthermore, the solid angles of both To and Tp indicate that
Table 1
Enantioselective hydroamination/cyclization of 2,2-diphenyl-4-penten-1-amine
catalyzed by To MgMe and To CaC(SiHMe ) .
2 3
ToT)
tBu
(
(
2
Mg and (Tp
2
) Mg are not sterically reasonable complexes
at least when both ligands are bonded in a tridentate fashion).
Thus, the moniker ’tetrahedral enforcer’, often applied to Tp , [1]
is also an appropriate descriptor of the steric properties of To .
T
T
H
N
tBu
Ph Ph
T
10 mol% catalyst
T
2
H N
Consistently, To MgMe is thermally robust in solution, and no
change in the 1H NMR spectra were observed after heating in
6 6
C D
Ph
ꢁ
T
toluene-d at 120 C for five days. Additionally, To MgMe can be
stored as a solid at room temperature in the absence of air and
moisture for extended times without observable decomposition.
8
Ph
T
A related calcium complex is accessible by reaction of H[To ] and
Entry Catalyst
R
Time
Temperature Conversion % ee
T
Ca(C(SiHMe
SiHMe and HC(SiHMe
can be removed under vacuum (Eq. (6)). On micromolar scale in
benzene-d , the reaction is complete after 5 min, but to insure
2
)
3
)
2
(THF)
2
[22] in benzene, which yields To CaC
T
1
2
3
To MgMe
Ph 24 h
Ph 12 h
Ph 5 min RT
RT
60
89%
ꢂ99%
ꢂ99%
0%
0%
0%
T
ꢁ
C
(
2
)
3
2
)
3
as a non-coordinating byproduct that
To MgMe
T
To CaC(SiHMe
2
)
3
6
complete conversion in larger scale preparations, the reaction was
allowed to stir for 1 h before workup.
T
T
AlthoughTo MgMeandTo CaC(SiHMe
)
2 3
are efficientcatalystsfor
the cyclization of 2,2-diphenyl-1-amino-pent-4-ene, the corre-
sponding pyrrolidine product is obtained as a racemic mixture (Table
1
). More promising results are obtained with C-(1-allyl-cyclohexyl)-
T
ꢁ
T
methylamine, which is cyclized by To MgMe at 60 C and by To CaC
SiHMe at room temperature to give the spiropyrrolidine product
in 36% and 18% ee, respectively (Table 2). Additionally, the substrate
(
2 3
)
T
2
,2-dimethyl-1-amino-pent-4-ene is cyclized by To MgMe with 27%
T
ee and by To CaC(SiHMe
2 3
) with 18% ee. Interestingly, hydro-
T
amination/cyclization of 2,2-dimethyl-1-amino-pent-4-ene and C-
The compound To CaC(SiHMe
2
)
3
is THF-free, and an intriguing
alkyl group is that it stabilizes
the formally four-coordinate calcium center through additional
T
(
1-allyl-cyclohexyl)-methylamine with To MgMe gives rise to the R
2 3
aspect of the eC(SiHMe )
T
pyrrolidines whereas To CaC(SiHMe
2 3
) givesrise totheS pyrrolidines.
1
We are currently working to elucidate the mechanism behind the
b
-agostic SiH interactions. Thus, the
.89 ppm and 153 Hz from the H NMR spectrum and the
dSiH and
J
SiH of values
T
T
1
2 3
stereo-defining step of To MgMe- and To CaC(SiHMe ) -catalyzed
4
n
SiH
ꢀ1
hydroamination/cyclization given this interesting observation.
However such differences in absolute configuration have also been
reported in rare earth metal complex-catalyzed hydroaminations
with scandium providing a different absolute configuration than
larger metal centers [24]. Thus, the substituents on the alkyl chain
appear to be the most important component of the aminoalkene for
stereoselectivity in these cyclizations.
values of 2106 and 1877 cm in the infrared spectrum indicate the
presence of -agostic SiH groups, as are observed in the starting
material Ca(C(SiHMe (THF) [22] These data are consistent with
a structure of To CaC(SiHMe containing one or two -agostic SiH
b
2
)
3
)
2
2
T
2
)
3
b
1
interactions that are fluxional on the H NMR timescale but
exchange slower than the IR timescale. The three oxazoline groups
1
are equivalent on the H NMR timescale. Additionally, no
n
CN bands
T
T
ꢀ
1
2 3
While the enantioselectivity of To MgMe and To CaC(SiHMe )
from 1630 to 1615 cm were detected that would correspond to
non-coordinated oxazoline. Only a single nCN band at 1569 cm
was detected that was assigned as the symmetric normal stretching
mode. These data indicate that all three oxazolines are coordinated
to calcium and remain coordinated, at least for timeframes longer
than the IR timescale.
ꢀ
1
are relatively low in comparison to the highly selective catalysts of
Hultzsch (Sc and Lu) [25] and Schafer (Zr) [26] these chiral oxa-
zolinylborato magnesium and calcium catalysts provide the highest
reported % ee’s for Group 2-catalyzed hydroamination/cyclization.
We are aware of only two reports of enantioselective hydro-
amination/cyclization with Group 2-based catalysts. Previously,
Hultzsch and co-workers described chiral binaphthyl-derived
tetraamine dimagnesium complexes that cyclize 2,2-diphenyl-1-
amino-pent-4-ene with 14% ee, though 2,2-dimethyl-1-amino-
pent-4-ene (4% ee) and C-(1-allyl-cyclohexyl)-methylamine (6% ee)
were less successful [7b]. Additionally, Harder and co-workers
2
.3. Hydroamination/cyclization of aminoalkenes
T
We investigated our new optically active complexes To MgMe
T
and To CaC(SiHMe
tion of the aminopentenes 2,2-diphenyl-1-amino-pent-4-ene,
,2-dimethyl-1-amino-pent-4-ene, and C-(1-allyl-cyclohexyl)-
methylamine as test substrates. The results are summarized in
2 3
) as catalysts for the hydroamination/cycliza-
2
Table 2
T
Enantioselective hydroamination/cyclization of C-(1-allyl-cyclohexyl)-methylamine
2 3
Tables 1e3. In general, the calcium compound To CaC(SiHMe )
catalyzes the cyclization of all three substrates at a significantly
greater rate than hydroaminations catalyzed by To MgMe.
While we expected that the presumed catalyst intermediate
T
T
2 3
catalyzed by To MgMe and To CaC(SiHMe ) .
T
NH
10 mol% catalyst
T
To MNHCH
2
CR
2
CH
2
CH]CH
2
(R
2
¼ Ph
2 2 5 2
, e(CH ) e, Me ) would be
2
H N
6 6
C D
more reactive for M ¼ Ca than M ¼ Mg based on size and related
rates of hydroamination/cyclizations catalyzed by diketiminate
T
calcium and magnesium complexes, [3] initiation of To CaC
Entry Catalyst
Tim
Temperature Conversion
% ee
(
(
SiHMe
SiHMe
2
)
)
3
is surprisingly fast given the unusual properties of the -C
group [22]. This alkyl ligand is both sterically encum-
bered and relatively non-basic at the central carbon, and as a result
Lewis acids such as B(C react with a peripheral SiH rather than
abstract the alkyl group in the typical fashion [23].
T
1
2
3
4
To MgMe
24 h RT
26 h 60
5 min RT
no conversion n.a.
2
3
T
ꢁ
ꢁ
To MgMe
C
C
93%
36% (R)
T
To CaC(SiHMe
2
)
)
3
3
ꢂ99%
18% (S)
n.d.
6
F
5
)
3
To CaC(SiHMe
T
(1% cat) 7 d
80
ꢃ10%
2

232
S.R. Neal et al. / Journal of Organometallic Chemistry 696 (2011) 228e234
1
1
Table 3
chemical shift scale by adding ꢀ381.9 ppm B NMR spectra were
29
Enantioselective hydroamination/cyclization of 2,2-dimethyl-4-penten-1-amine
catalyzed by To MgMe and To CaC(SiHMe ) .
2 3
3 2
referenced to an external sample of BF $Et O and Si NMR spectra
T
T
were referenced to an external sample of tetramethylsilane. Accu-
rate mass ESI mass spectrometry was performed using the Agilent
QTOF 6530 equipped with the Jet Stream ESI source. An Agilent ESI
test mix was used for tuning and calibration. Accurate mass data
was obtained in the positive ion mode using a reference standard
with ions at 121.05087 and 922.00979. The mass resolution
(FWHM) was maintained at 18,000.
H
N
1
0 mol% catalyst
2
H N
6 6
C D
Catalyst
T
Time
Temperature
Conversion
% ee
1
2
3
To MgMe
To MgMe
To CaC(SiHMe
7 d
5 d
5 min
RT
80
RT
none
80%
100%
n.a.
27% (R)
18% (S)
T
ꢁ
4.1.1. 2Li-4-tert-butyl-2-oxazoline [LiOxtBu
]
C
T
2
)
3
Hexamethyldisilazane (1.0 mL, 4.80 mmol) was added to
a Schlenk flask followed by 15 mL of THF. The solution was cooled
ꢁ
to ꢀ78 C then n-butyllithium (1.92 mL, 4.80 mmol) was slowly
reported that bis(oxazoline) disilazidocalcium catalyzes the cycli-
zation of 2,2-diphenyl-1-amino-pent-4-ene in up to 10% ee [7a].
added. The solution was allowed to stir for 45 min, and then a THF
(5 mL) solution of degassed 4S-tert-butyl-2-oxazoline was added
slowly via cannula. The solution immediately turned bright
ꢁ
yellow. The yellow solution was stirred at ꢀ78 C for 1 h then
3
. Conclusion
warmed to room temperature and allowed to stir for 2 h. All
volatiles were removed in vacuo. The resulting yellow solid was
dissolved in warm hexane and cooled to ꢀ30 C. The resulting off-
The contrasting stereoselectivitybetweentheneutralGroup2 tris
ꢁ
(
oxazolinyl)borate catalyzed hydroamination/cyclization described
here and the dicationic trisox scandium alkyls that polymerize
-olefins with high stereoregularity is striking. Both reactions are
presumed to involve insertion of an olefin into an MeE bond, albeit
Mg/CaeN versus SceC bonds and intramolecular cyclization versus
a chain and/or ligand mediated intramolecular process. Clearly, the
factors that control stereoselective insertion in olefinpolymerization
are significantly different than the factors that influence cyclization
of aminoalkenes in hydroamination reactions. A related comparison
ꢁ
white solid was isolated by filtration at ꢀ30 C and dried under
tBu
vacuum to yield 0.461 g (3.46 mmol, 76%) of crude LiOx
that is
a
1H NMR
contaminated with small amounts of HN(SiMe ) .
3 2
2
(
400 MHz, tetrahydrofuran-d
8
):
d
4.02 (br d,
O), 3.74 (t, JHH ¼ 9.6 Hz, 1 H, LiCNCHCMe
HH ¼ 7.2 Hz, 1 H, LiCNCHCMe CH O), 0.95 (s, 9 H,
O). C{ H} NMR (100 MHz, tetrahydrofuran-d ):
156.98 (LiCNCHCMe CH O), 73.54 (LiCNCHCMe CH O), 65.64
LiCNCHCMe CH O), 33.48 (LiCNCHCMe
O). Li NMR (156 MHz, benzene-d
970 s, 2908 m, 2873 m, 2833 m, 2718 w, 2000 w (
opened isocyanide), 1635 w ( CN; oxazolide), 1479 m, 1465 m, 1400
w, 1367 m, 1336 w, 1243 w, 1222 w, 1190 w, 1020 w, 942 w, 922 w,
J
HH ¼ 8.4 Hz, 1 H,
3
LiCNCHCMe
.23 (br d,
LiCNCHCMe
3
2
CH
J
2
3
CH O),
2
3
3
2
13
1
3
CH
2
8
d
3
2
3
2
(
Me
3
2
3
CH
):
2
d
O), 27.27 (LiCNCHC-
was considered by Marks and co-workers, who investigated C
symmetric and chiral ansa-lanthanoidocene complexes Me
Me )(C R*)LnR as catalysts for enantioselective hydrogenation
1
-
7
ꢀ1
3
CH
2
6
1.22. IR (KBr, cm ):
2
Si
2
nCN; ring-
(
C
5
4
5 3
H
n
and hydroamination/cyclization [27]. In those cases, good enantio-
selectivities were obtained for both reactions (up to 74% for hydro-
amination and upto 96%forhydrogenation), and olefin insertionwas
proposed as the stereochemistry defining step for both catalytic
transformations.
820 w, 779 w.
.1.2. Li[ToT]
4
A solution of 4S-tert-butyl-2-oxazoline (2.77 g, 21.8 mmol) in
ꢁ
THF (100 mL) was cooled to ꢀ78 C in a dry ice/acetone bath. tert-
4
. Experimental
Butyllithium (13.6 mL, 23.1 mmol, 1.7 M) was added in a dropwise
fashion, and the solution turned from colorless to bright yellow. The
ꢁ
4.1. General procedures
reaction mixture was stirred for 30 min at ꢀ78 C, and then PhBCl
2
(
0.88 mL, 6.78 mmol) was slowly added. The yellow solution was
All reactions were performed under a dry argon atmosphere
allowed to slowly warm to room temperature and was then stirred
for 72 h. The volatiles were evaporated, and the resulting yellow
using standard Schlenk techniques or under a nitrogen atmosphere
in a glovebox unless otherwise indicated. Dry, oxygen-free solvents
were used throughout. Benzene, toluene, pentane and tetrahy-
drofuran were degassed by sparging with nitrogen, filtered through
T
solid was extracted with diethyl ether to yield crude Li[To ] (2.23 g,
T ꢀ
4.71 mmol, 65%). The most efficient way to obtain pure [ToT ] is to
T
T
1
11
transform Li[To ] into H[To ], but H NMR and B NMR spectros-
T
1
activated alumina columns, and stored under N
toluene-d and tetrahydrofuran-d were vacuum transferred from
Na/K alloy and stored under in the glovebox. Ca(C
SiHMe (THF) [22] Me Mg$(O [28] 2,2-diphenyl-4-
2
. Benzene-d
6
,
copy are useful for verifying complete conversion to Li[To ]. H NMR
3
8
8
(methanol-d
4
):
d
7.42 (d,
J
HH ¼ 5.6 Hz, 2 H, ortho-C
6 5
H ), 7.02
3
3
N
2
(t, JHH ¼ 7.2 Hz, 2 H, meta-C
6
H
5
), 6.94 (t, JHH ¼ 7.2 Hz, 1 H, para-
3
3
(
2
)
3
)
2
2
2
2
C
H
4 8
)
2
C
6
H
5
), 3.89 (d, JHH ¼ 8.0 Hz, 6 H, OCH
2
7
), 3.75 (t, JHH ¼ 8.4 Hz, 3 H,
penten-1-amine [29] 2,2-dimethyl-4-penten-1-amine [30] and
C-(1-allyl-cyclohexyl)-methylamine [31] were prepared by pub-
lished procedures. All the aminoalkenes were degassed and stored
with 4 Å molecular sieves in a glovebox prior to use. All other
NCHCMe
trile-d ):
3
d
), 0.86 (s, 27 H, NCHCMe
3
). Li NMR (156 MHz, acetoni-
1
1
3
0.11. B NMR (128 MHz, methanol-d
4
):
d
ꢀ17.0.
4.1.3. H[ToT]
Crude Li[To ] (1.316 g, 2.78 mmol) was dissolved in methylene
chloride (50 mL). [HNEt ]Cl (0.455 g, 3.31 mmol) was added, and
1
11
13
1
T
chemicals used here are commercially available. H, B, C{ H},
2
9
1
and Si{ H} NMR spectra were collected on a Bruker DRX-400
spectrometer or a Bruker Avance II 700 spectrometer with a Bruker
3
the resulting yellow suspension was stirred overnight at room
temperature. All volatile materials were evaporated under reduced
pressure. The resulting pale yellow solid was redissolved in
a minimal amount of methylene chloride and passed through
a plug of grade III neutral alumina (16 mm ꢄ 76 mm) using
methylene chloride (100 mL) as the eluent. The solvent was
Z-gradient inverse TXI 1H/ C/ N 5 mm cryoprobe. C{ H} NMR
13
15
13
1
chemical shifts of boron-bonded carbons were not observed in
11
15
several cases due to the quadrupolar B nucleus. N chemical
shifts were also determined on the Bruker Avance II 700 spec-
1
15
15
trometer by He N HMBC experiments; N chemical shifts were
originally referenced to liquid NH and recalculated to the CH NO
T
3
3
2
removed under vacuum to yield a pale yellow solid. The solid H[To ]

S.R. Neal et al. / Journal of Organometallic Chemistry 696 (2011) 228e234
233
was extracted with benzene (3 ꢄ 15 mL) to remove residual [HNEt
3
]
25.84 (NCHCMe
(128 MHz, benzene-d
3
), 4.18 (C(SiHMe
2
)
3
), 3.73 (C(SiHMe
2
)
3
). 11B NMR
1
5
Cl. The filtrate was collected, and the solvent was removed under
vacuum to yield a pale yellow solid. This crude product was purified
by silica gel chromatography (16 mm ꢄ 140 mm, hexane:
6
):
d
ꢀ16.7. N NMR (70.9 MHz, benzene-d
6
):
2
9
1
d
ꢀ163.7. Si{ H} NMR (79.5 MHz, benzene-d
6
):
d
ꢀ20.4. IR (KBr,
ꢀ1
cm ): 3044 w, 3074 w, 2958 m, 2903 m, 2870 w, 2106 m (SiH),
EtOAc:NEt
3
¼ 20:6:1, R
f
¼ 0.27) to afford 0.714 g (1.53 mmol, 55%) of
1877 w (SiH), 1569 m (C]N), 1478 w, 1254 w. Anal. Calcd for
T
hydrogen tris(4S-tert-butyl-2-oxazolinyl)phenylborate (H[To ]). H
C
34
H
62BCaN
3
O
3
Si
3
(HC(SiHMe
2
)
3
): C, 55.55; H, 9.55; N, 4.74. Found:
T
ꢁ
[
To ] was further dried over P
2
O
5
in benzene without loss of yield.
C, 55.67; H, 9.44; N, 4.61. Mp 226e227 C (dec).
1
3
H NMR (400 MHz, benzene-d
6
):
), 7.46 (t, JHH ¼ 7.6 Hz, 2 H, meta-C
), 3.78 (m, 6 H, OCH
), 3.44 (dd, JHH ¼ 10.2 Hz, 7.6 Hz, 3
), 0.80 (s, 27 H, CMe
134.99 (ortho-C ), 127.88 (meta-C
), 73.18 (NCHCMe ), 39.16 (OCH ), 33.86 (CMe
). B NMR (128 MHz, benzene-d
70.9 MHz, benzene-d ):
ꢀ174.5. IR (KBr, cm ): 3069 w, 3045 w,
955 s, 2901 m, 2869 m, 1601 s (C]N), 1478 m, 1423 m, 1392 w,
d
8.12 (d, JHH ¼ 7.2 Hz, 1 H, ortho-
3
3
C
H
6 5
6
H
5
), 7.25 (t, JHH ¼ 7.2 Hz, 2
4.1.6. General conditions for hydroamination/cyclization
T
T
H, para-C
H, NCHCMe
benzene-d ):
6
H
5
2
In a glovebox, To MgMe or To CaC(SiHMe
lent) and aminoalkene (10 equivalents) were massed in separate
test tubes. The catalyst was dissolved in benzene-d and transferred
to a test tube containing the aminoalkene. This solution was either
added to a dry NMR tube and capped with a septa for room
temperature reactions or added to a dry NMR tube fitted with
2 3
) catalyst (1 equiva-
13
1
3
3
). C{ H} NMR (175 MHz,
), 126.36
), 26.14
6
d
6
H
5
6
H
5
6
(
(
(
para-C
6
H
5
3
2
3
1
1
15
CMe
3
6
):
d
ꢀ16.5. N NMR
ꢀ
1
6
d
1
2
a J-Young valve for reactions at elevated temperatures. H NMR
1
for
362 w, 1208 w, 1176 w, 969 m. MS (ESI) exact mass Calculated
spectra were taken at regular intervals.
þ
C
27
H
42BN
3
O
3
:
m/e 468.3392 ([M ]), Found: 468.3398
ꢁ
(D
ꢀ1.29 ppm). Mp 98e102 C.
4.1.7. Determination of % ee for cyclohexyl- and dimethyl-
pyrrolidine
T
4.1.4. To MgMe
The NMR sample was transferred to a flask and all volatiles were
vacuum transferred under reduced pressure. The solution is then
transferred to an NMR tube. The amount of pyrrolidine is calculated
T
A yellow benzene solution of H[To ] (0.441 g, 0.943 mmol) was
slowly added to a rapidly stirring suspension of Me Mg$(O
0.241 g, 1.04 mmol) in benzene at room temperature. Vigorous
bubbling was observed upon addition. After addition was
complete, the suspension was stirred for 2 h; excess Me Mg$
, which is insoluble under reaction conditions, was
2
2 4 8 2
C H )
1
(
from the H NMR spectrum using tetrakis(trimethylsilyl)silane
as an internal standard. Hünig’s base (2 equivalents) and (S)-
(þ)-Mosher’s chloride (1.2 equivalents) were added to the NMR
tube. After 20 min, the NMR sample was added to a vial and all
volatiles were removed in vacuo. The pyrrolidine-Mosher amide
2
2 4 8 2
(O C H )
removed by filtration. The filtrate was evaporated under reduced
pressure, and the resulting solid was washed with pentane to
yield To MgMe (0.384 g, 0.759 mmol, 80.5%). X-ray quality crystals
are obtained by cooling a concentrated toluene solution of
was extracted with pentane (3 ꢄ 2 mL) and the volatiles were
T
19
removed. The % ee was then determined by integration of the
F
ꢁ
NMR spectrum at 60 C in chloroform-d.
T
ꢁ
1
To MgMe to ꢀ78 C. H NMR (700 MHz, benzene-d
6
): d 8.22 (d,
HH ¼ 7.7 Hz, 2 H, meta-
3
3
J
HH ¼ 7.7 Hz, 2 H, ortho-C
6
H
5
), 7.52 (t,
), 7.33 (t, JHH ¼ 7.4 Hz, 1 H, para-C
HH ¼ 5.6 Hz, 3 H, OCH
J
4.1.8. Determination of % ee for diphenyl-pyrrolidine
3
2
C
J
(
6
H
5
6
H
5
), 3.72 (dd, JHH ¼ 9.8 Hz,
The NMR sample was transferred to a small flask and the 4,4-
3
2
), 3.59 (v t, JHH ¼ 9.8 Hz, 3 H, OCH
HH ¼ 5.6 Hz, 3 H, NCHCMe ), 0.72 (s, 27 H,
):
), 127.23
), 70.28 (OCH ),
2
), 3.45
diphenyl-2-methylpyrrolidine product was vacuum distilled using
2
3
ꢁ
ꢀ6
dd,
CMe
193.76 (br, B-CNCHCMe
meta-C ), 126.16 (para-C
), 26.28 (CMe
J
HH ¼ 9.8 Hz,
J
3
a Kugelrohr (w120 C,10 torr). The distillate was then transferred
to an NMR tube with chloroform-d. The amount of pyrrolidine
13
1
3
), ꢀ0.65 (s, 3 H, MgMe). C{ H} NMR (175 MHz, benzene-d
6
1
d
3
CH
2
O), 136.44 (ortho-C
6
H
5
product was calculated from the H NMR spectrum using tetrakis
(
3
6
H
5
6
H
5
), 73.59 (NCHCMe
3
2
(trimethylsilyl)silane as an internal standard. Hünig’s base
(2 equivalents) and (S)-(þ)-Mosher’s chloride (1.2 equivalents)
were added to the NMR tube. After 20 min, all volatiles were
removed in vacuo. The amide product was extracted with pentane
(3 ꢄ 2 mL) and the volatiles were removed. The % ee was then
11
4.11 (CMe
3
3
), ꢀ13.69 (MgMe). B NMR (128 MHz,
ꢀ17.1. N NMR (70.9 MHz, benzene-d ): ꢀ178.2.
IR (KBr, cm ): 3045 w, 2958 m, 2869 m, 1585 s (C]N), 1478 m,
396 w, 1365 m, 1196 s, 966 m. Anal. Calcd for C28 44BMgN (-
): C, 64.72; H, 8.83; N, 7.08. Found: C, 64.44; H, 8.87; N,
15
benzene-d
6
):
d
6
d
ꢀ
1
1
H
3 3
O
1
C
7
4
H
8
O
2
determined by integration of the H NMR spectrum at ambient
ꢁ
.52. Mp 238e239 C (dec).
temperature in chloroform-d.
T
4
.1.5. To CaC(SiHMe
2
)
3
Acknowledgement
Ca(C(SiHMe (THF) (0.086 g, 0.153 mmol) was dissolved in
2
)
3
)
2
2
T
benzene (5 mL). In a separate vial, H[To ] (0.054 g, 0.116 mmol) was
dissolved in benzene (5 mL) and was added to the Ca(C(SiH-
The U.S. Department of Energy, Office of Basic Energy Science
(DE-AC02-07CH11358) is acknowledged for partial financial
support of this work for A.D. Sadow. The ACS Green Chemistry
Institute-PRF provided material support. Aaron D. Sadow is an
Alfred P. Sloan Research Fellow.
Me
2 3 2 2
) ) (THF) solution along with an additional 5 mL of benzene.
The reaction mixture was stirred for 1 h, and then all the volatiles
T
were removed in vacuo to yield an orange solid. To CaC(SiHMe
was extracted with pentane; the yellow pentane solution was
placed under vacuum overnight to remove the HC(SiHMe
2 3
)
2
)
3
Appendix A. Supplementary material
byproduct to yield a pale yellow solid (49.2 mg, 0.071 mmol, 61%).
1
3
H NMR (400 MHz, benzene-d
6
):
d
8.11 (d, JHH ¼ 7.2 Hz, 2 H, ortho-
CCDC 787151 contains the supplementary crystallographic data
for this paper. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
3
3
C
6
H
5
), 7.49 (t, JHH ¼ 7.2 Hz, 2 H, meta-C
6
H
5
), 7.30 (t, JHH ¼ 7.2 Hz, 1
1
3
H, para-C
H
6 5
), 4.89 (d, sept, JSiH ¼ 153 Hz, JHH ¼ 3.2 Hz, 3 H, SiH),
2 3
3
.72 (dd,
J
HH ¼ 9.2 Hz,
J
HH ¼ 4 Hz, 3 H, OCH
HH ¼ 9.2 Hz, JHH ¼ 4 Hz, 3 H, NCHCMe ), 3.48 (v t, JHH ¼ 9.2 Hz, 3
H, OCH ), 0.73 (s, 27 H, NCHCMe ), 0.52 (d,
SiHCH ), 0.50 (d,
HH ¼ 3.2 Hz, 9 H, SiHCH
benzene-d ): 135.97 (ortho-C ), 126.83 (meta-C
para-C ), 74.15 (NCHCMe ), 68.91 (OCH ), 34.02 (NCHCMe
2
), 3.67 (dd,
3
3
J
3
3
2
3
J
HH ¼ 3.2 Hz, 9 H,
Appendix. Supplementary data
3
13
1
3
J
3
). C{ H} (100 MHz,
), 125.59
),
6
d
6
H
5
6
H
5
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.jorganchem.2010.08.057.
(
6
H
5
3
2
3

234
S.R. Neal et al. / Journal of Organometallic Chemistry 696 (2011) 228e234
References
[13] (a) S. Bellemin-Laponnaz, L.H. Gade, Chem. Commun. (2002) 1286e1287;
b) S. Bellemin-Laponnaz, L.H. Gade, Angew. Chem. Int. Ed. 41 (2002)3473e3475.
14] (a) B.D. Ward, S. Bellemin-Laponnaz, L.H. Gade, Angew. Chem. Int. Ed. 44
(
[
[1] (a) R. Han, G. Parkin, Organometallics 10 (1991) 1010e1020;
(
(
2005) 1668e1671;
(b) R. Han, A. Looney, G. Parkin, J. Am. Chem. Soc. 111 (1989) 7276e7278;
(c) R. Han, G. Parkin, J. Organomet. Chem. 292 (1990) C43eC46;
(d) R. Han, G. Parkin, J. Am. Chem. Soc. 114 (1992) 748e757.
b) L. Lukesova, B.D. Ward, S. Bellemin-Laponnaz, H. Wadepohl, L.H. Gade,
Dalton Trans. (2007) 920e922;
c) L. Lukesova, B.D. Ward, S. Bellemin-Laponnaz, H. Wadepohl, L.H. Gade,
(
[
[
2] M.H. Chisholm, N.W. Eilerts, J.C. Huffman, S.S. Iyer, M. Pacold, K. Phomphrai,
J. Am. Chem. Soc. 122 (2000) 11845e11854.
3] (a) M.R. Crimmin, I.J. Casely, M.S. Hill, J. Am. Chem. Soc. 127 (2005) 2042e2043;
Organometallics 26 (2007) 4652e4657.
[
[
15] D. Mukherjee, A. Ellern, A.D. Sadow, J. Am. Chem. Soc. 132 (2010) 7582e7583.
16] B. Baird, A.V. Pawlikowski, J. Su, J.W. Wiench, M. Pruski, A.D. Sadow, Inorg.
Chem. 47 (2008) 10208e10210.
(
b) M.R. Crimmin, M. Arrowsmith, A.G.M. Barrett, I.J. Casely, M.S. Hill,
P.A. Procopiou, J. Am. Chem. Soc. 131 (2009) 9670e9685.
[17] (a) L.A. Silks, J. Peng, J.D. Odom, R.B. Dunlap, J. Org. Chem. 56 (1991)
[
[
4] For recent reviews of calcium mediated hydroamination see
6
733e6736;
b) M. Ollivault-Shiflett, D.B. Kimball, L.A. Silks, J. Org. Chem. 69 (2004)
150e5152.
18] (a) C. Mazet, V. Kohler, A. Pfaltz, Angew. Chem. Int. Ed. 44 (2005) 4888e4891;
b) V. Kohler, C. Mazet, A. Toussaint, K. Kulicke, D. Haussinger, M. Neuburger,
(
(
a) S. Harder, Chem. Rev. (2010) 3852e3876;
b) A.G.M. Barrett, M.R. Crimmin, M.S. Hill, P.A. Procopiou, Proc. R. Soc. (2010)
(
5
927e963.
[
5] (a) S. Harder, J. Brettar, Angew. Chem. Int. Ed. 45 (2006) 3474e3478;
(
(
(
b) J. Spielmann, S. Harder, Eur. J. Inorg. Chem. (2008) 1480e1486;
c) M.R. Crimmin, A.G.M. Barrett, M.S. Hill, P.B. Hitchcock, P.A. Procopiou,
S. Schaffner, S. Kaiser, A. Pfaltz, Chem. Eur. J. 14 (2008) 8530e8539.
19] W.R. Leonard, J.L. Romine, A.I. Meyers, J. Org. Chem. 56 (1991) 1961e1963.
20] (a) D. White, B.C. Taverner, P.G.L. Leach, N.J. Coville, J. Comput. Chem. 14
[
[
Organometallics 27 (2008) 497e499;
d) A.G.M. Barrett, M.R. Crimmin, M.S. Hill, P.B. Hitchcock, P.A. Procopiou,
Dalton Trans. (2008) 4474e4481;
e) A.G.M. Barrett, M.R. Crimmin, M.S. Hill, P.B. Hitchcock, S.L. Lomas,
M.F. Mahon, P.A. Procopiou, K. Suntharalingam, Organometallics 27 (2008)
300e6306.
6] (a) I. Nieto, F. Cervantes-Lee, J.M. Smith, Chem. Commun. (2005) 3811e3813;
b) M. Arrowsmith, A. Heath, M.S. Hill, P.B. Hitchcock, G. Kociok-Köhn,
Organometallics 28 (2009) 4550e4559;
c) M. Arrowsmith, M.S. Hill, C. Kociok-Köhn, Organometallics 28 (2009)
730e1738.
7] (a) F. Buch, S. Harder, Z. Naturforsch 63b (2008) 169e177;
b) P. Horrillo-Martínez, K.C. Hultzsch, Tetrahedron Lett. 50 (2009) 2054e2056.
(
(
(
1993) 1042e1049;
b) D. White, N.J. Coville, Adv. Organomet. Chem. (1994) 95e158 Academic
(
Press: New York.
[
[
21] (a) I.A. Guzei, M. Wendt, “Solid-G” UW-Madison, WI, USA, 2004;
6
(
b) I.A. Guzei, M. Wendt, Dalton Trans. (2006) 3991e3999.
22] K. Yan, B.M. Upton, A. Ellern, A.D. Sadow, J. Am. Chem. Soc. 131 (2009)
5110e15111.
[
[
(
1
[
[
[
23] X. Yang, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 116 (1994) 10015e10031.
24] X. Yu, T.J. Marks, Organometallics 26 (2007) 365e376.
25] D.V. Gribkov, K.C. Hultzsch, F. Hampel, J. Am. Chem. Soc. 128 (2006)
(
1
3748e3759.
(
[
[
26] M.C. Wood, D.C. Leitch, C.S. Yeung, J.A. Kozak, L.L. Schafer, Angew. Chem. Int.
Ed. 46 (2007) 354e358.
27] M.A. Giardello, V.P. Conticello, L. Brard, M.R. Gagne, T.J. Marks, J. Am. Chem.
Soc. 116 (1994) 10241e10254.
[
[
8] J.F. Dunne, J. Su, A. Ellern, A.D. Sadow, Organometallics 27 (2008) 2399e2401.
9] (a) A.V. Pawlikowski, T.S. Gray, G. Schoendorff, B. Baird, A. Ellern, T.L. Windus,
A.D. Sadow, Inorg. Chim. Acta 362 (2009) 4517e4525;
(
(
b) A.V. Pawlikowski, A. Ellern, A.D. Sadow, Inorg. Chem. 48 (2009) 8020e8029;
c) J.F. Dunne, K. Manna, J.W. Wiench, A. Ellern, M. Pruski, A.D. Sadow, Dalton
[
[
28] B. Rickborn, J. Baranski, D. Tobia, J. Org. Chem. 54 (1989) 4253e4256.
29] S. Hong, S. Tian, M.V. Metz, T.J. Marks, J. Am. Chem. Soc. 125 (2003)
Trans. 29 (2010) 641e653.
10] (a) M.R. Gagne, T.J. Marks, J. Am. Chem. Soc. 111 (1989) 4108e4109;
b) M.R. Gagne, C.L. Stern, T.J. Marks, J. Am. Chem. Soc. 114 (1992) 275e294.
11] D.M. Tellers, S.J. Skoog, R.G. Bergman, T.B. Gunnoe, W.D. Harman, Organo-
metallics 19 (2000) 2428e2432.
1
4768e14783.
30] (a) P.H. Martínez, K.C. Hultzsch, F. Hampel, Chem. Commun. (2006)
221e2223;
b) C.F. Bender, R.A. Widenhoefer, J. Am. Chem. Soc. 127 (2005) 1070e1071.
31] D. Riegert, J. Collin, A. Meddour, E. Schulz, A. Trifonov, J. Org. Chem. 71 (2006)
514e2517.
[
[
[
[
(
2
(
[
12] H.-A. Ho, J.F. Dunne, A. Ellern, A.D. Sadow, Organometallics 29 (2010)
2
4105e4114.