σ-Insertive mechanism versus concerted non-insertive mechanism in the intramolecular hydroamination of aminoalkenes catalyzed by phenoxyamine magnesium complexes: A synthetic and computational study

Article abstract of DOI:10.1002/chem.201406468

The phenoxyamine magnesium complexes [{ONN}MgCH₂Ph] (4 a: {ONN}=2,4-tBu₂-6-(CH₂NMeCH₂CH₂NMe₂)C₆H₂O^-; 4 b: {ONN}=4-tBu-2-(CH₂NMeCH₂CH

Full text of DOI:10.1002/chem.201406468

DOI: 10.1002/chem.201406468
Full Paper
&
Reaction Mechanisms |Hot Paper|
s-Insertive Mechanism versus Concerted Non-insertive Mechanism
in the Intramolecular Hydroamination of Aminoalkenes Catalyzed
by Phenoxyamine Magnesium Complexes: A Synthetic and
Computational Study
Xiaoming Zhang,[a] Sven Tobisch,*[b] and Kai C. Hultzsch*[a, c]
Abstract: The phenoxyamine magnesium complexes
[{ONN}MgCH2Ph] (4a: {ONN}=2,4-tBu2-6-(CH2NMeCH2CH2-
NMe2)C6H2OÀ; 4b: {ONN}=4-tBu-2-(CH2NMeCH2CH2NMe2)-6-
(SiPh3)C6H2OÀ) have been prepared and investigated with re-
spect to their catalytic activity in the intramolecular hydro-
amination of aminoalkenes. The sterically more shielded tri-
phenylsilyl-substituted complex 4b exhibits better thermal
stability and higher catalytic activity. Kinetic investigations
using complex 4b in the cyclisation of 1-allylcyclohexyl)me-
thylamine (5b), respectively, 2,2-dimethylpent-4-en-1-amine
(5c), reveal a first-order rate dependence on substrate and
catalyst concentration. A significant primary kinetic isotope
effect of 3.9Æ0.2 in the cyclisation of 5b suggests signifi-
cant NÀH bond disruption in the rate-determining transition
state. The stoichiometric reaction of 4b with 5c revealed
that at least two substrate molecules are required per mag-
nesium centre to facilitate cyclisation. The reaction mech-
anism was further scrutinized computationally by examina-
tion of two rivalling mechanistic pathways. One scenario in-
volves a coordinated amine molecule assisting in a concerted
non-insertive NÀC ring closure with concurrent amino
proton transfer from the amine onto the olefin, effectively
combining the insertion and protonolysis step to a single
step. The alternative mechanistic scenario involves a reversi-
ble olefin insertion step followed by rate-determining proto-
nolysis. DFT reveals that a proton-assisted concerted NÀC/
CÀH bond-forming pathway is energetically prohibitive in
comparison to the kinetically less demanding s-insertive
pathway (DDG° =5.6 kcalmolÀ1). Thus, the s-insertive path-
way is likely traversed exclusively. The DFT predicted total
barrier of 23.1 kcalmolÀ1 (relative to the {ONN}Mg pyrrolide
catalyst resting state) for magnesiumÀalkyl bond aminolysis
matches the experimentally determined Eyring parameter
(DG° =24.1(Æ0.6) kcalmolÀ1 (298 K)) gratifyingly well.
Group 8–12[8]) in the last two decades. Main group metal-
based catalysts have been studied more intensely only recent-
ly. For example, alkali metal-based hydroamination catalysts
have been known since the 1950s,[9] but more detailed studies
have only begun to emerge in recent years.[10] Alkaline-earth
metal complexes[11] are also quite active catalysts for the
hydroamination of alkenes[12,13] thanks to the similarity be-
tween the chemistry of alkaline-earth metals and that of the
rare-earth metals. Unfortunately, alkaline-earth metal com-
plexes are prone to facile Schlenk-type ligand redistribu-
tions[11,14] that have hampered the development of chiral alka-
line-earth metal hydroamination catalysts for enantioselective
hydroamination reactions.[15] However, we have recently devel-
Introduction
Nitrogen-containing compounds represent valuable and indus-
trially important bulk chemicals, specialty chemicals and
pharmaceuticals.[1] The hydroamination reaction[2] offers facile
access to amines, enamines and imines in a waste-free, highly
atom-efficient and green manner. Research efforts have focus-
ed primarily on the development of transition metal-based
catalyst systems (early transition metals of Group 3–5 including
the lanthanides,[3–6] actinides,[7] and late transition metals of
[a] Dr. X. Zhang, Prof. Dr. K. C. Hultzsch
Department of Chemistry and Chemical Biology, Rutgers
The State University of New Jersey, 610 Taylor Road
Piscataway, NJ 08854-8087 (USA)
oped
phenoxyamine
{ONN}
magnesium
complexes
(Figure 1)[13,16,17] that resist ligand redistribution and a chiral
catalyst, complex 2, which achieved enantioselectivities of up
to 93% ee in the intramolecular hydroamination of amino-
alkenes.[13b]
[b] Dr. S. Tobisch
University of St. Andrews, School of Chemistry
Purdie Building, North Haugh, St. Andrews, KY16 9ST (UK)
E-mail: st40@st-andrews.ac.uk
[c] Prof. Dr. K. C. Hultzsch
The alkaline-earth metal-catalysed hydroamination reaction
is generally believed to proceed in a similar manner as has
Present address: University of Vienna, Faculty of Chemistry
Institute of Chemical Catalysis, Währinger Strasse 38, 1090 Vienna (Austria)
E-mail: kai.hultzsch@univie.ac.at
been proposed for rare-earth metal-based catalysts.[3c,18]
A
metal amide species A undergoes migratory olefin insertion,
followed by protonolysis of the metal alkyl intermediate B to
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201406468.
Chem. Eur. J. 2015, 21, 7841 – 7857
7841
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 2. Proposal for a concerted non-insertive NÀC ring closure with con-
current amino proton transfer onto the olefin in the hydroamination/cyclisa-
tion (RR’NH=substrate or hydroamination product).
nesium[20] and
a cyclopentadienyl-bis(oxazolinyl)borate yt-
trium[18h] catalyst system indicate that the reaction proceeds
via a reversible migratory insertion step followed by rate-deter-
mining protonolysis.
In the present study we will scrutinize two rivalling mech-
anistic pathways for the intramolecular hydroamination of
aminoalkenes utilizing novel phenoxyamine {ONN} magnesium
complexes. One working hypothesis was that a coordinated
amine molecule could assist in a concerted non-insertive NÀC
ring closure with concurrent amino proton transfer from the
amine onto the olefin (Scheme 2), effectively combining the in-
sertion and protonolysis step to a single step. An alternative
scenario implies a reversible olefin insertion step and a rate-de-
termining protonolysis step (Scheme 3).
Figure 1. Previously studied phenoxyamine {ONN} magnesium hydro-
amination catalysts.[13, 17]
regenerate the catalytically active metal amide species A
(Scheme 1). Rare-earth metal-based catalyst systems are com-
monly thought to involve a rate-determining olefin insertion
step and fast protonolysis, resulting in zeroth order rate de-
pendency on substrate concentration. The relative rate of the
two steps is less clear for catalysts based on alkaline-earth
metals, because these catalysts predominantly,[12f,m–o,w, 13b] with
a few exceptions,[13a] display a first-order rate dependence on
substrate concentration, which is contradictory to a fast proto-
nolysis step.
Scheme 3. Proposal for an alternative mechanistic scenario with reversible
olefin insertion and rate-determining protonolysis.
Results and Discussion
Synthesis and characterization of model systems
Scheme 1. Postulated general mechanism for hydroamination/cyclisation of
aminoalkenes catalysed by alkali, alkaline-earth and rare-earth metals.
In accordance to a previous study on a chiral phenoxydiamine
magnesium catalyst,[13b] the related achiral magnesium com-
plexes 4a and 4b incorporating an ethylenediamine sidearm
can be most conveniently prepared via alkane elimination by
treating the appropriate phenoldiamine proligands 3a,
respectively, 3b, with Mg(CH2Ph)2(THF)2 in a 1-to-1 molar ratio
at room temperature (Scheme 4). The di-tert-butyl-substituted
complex 4a proved to be rather unstable and attempts to iso-
late complex 4a in solid form failed, due to formation of
a ligand redistribution product upon removal of solvent. There-
fore, complex 4a was generally prepared in situ and was used
directly for catalytic experiments. Complex 4b was obtained as
a white foam which is partially soluble in hexanes. Unfortu-
nately, all efforts to obtain X-ray quality crystals have been un-
successful so far. Complex 4b is stable at room temperature in
the solid state for several hours; however, slow decomposition
was observed over 3 days.
However, the observation of a significant primary kinetic iso-
tope effect[3c,w, z,4d,5s–v,12m–o,w, 13] and isotopic perturbation of
stereoselectivity[3c,w, 5v] in a number of catalyst systems based
on various metals are indicative of significant NÀH bond dis-
ruption in the transition state of the stereo- and rate-determin-
ing alkene insertion step. A plausible explanation for these
contradictory observations involves partial proton transfer
from a coordinated amine to the terminal olefin carbon of the
aminoalkene substrate in the ring-closing transition state
(Scheme 2).[3c] Similar concerted reaction mechanisms have
been proposed for a number of alkaline-earth[12m,n] and Group
4[5t–v] metal catalysts. A DFT study[19] performed on a bis(ureate)-
zirconium metal catalyst system[5u] has indeed confirmed this
proposal, while a study on a tris(oxazolinyl)phenylborate mag-
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7842
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 4. Synthesis of phenoxyamine {ONN} magnesium complexes.
1
The H NMR spectrum of complex 4b exhibits fluxional be-
haviour for the ethylenediamine sidearm similar to that ob-
served previously for the phenoxytriamine complex 1b.[13a]
Catalytic hydroaminations
Figure 2. Time dependence of substrate concentration in the hydroamina-
tion/cyclisation of (1-allylcyclohexyl)methylamine (5b, normalized to
C0 =0.167 molLÀ1) in [D6]benzene at 258C using varying concentrations of
4b. The error of measurement for the data points is estimated to be Æ5%
The catalytic activity of complexes 4a and 4b was studied in
the intramolecular hydroamination of a range of aminoalkene
substrates and can be compared to previously investigated
phenoxyamine magnesium catalysts (Table 1).
1
based on the accuracy of integrating signals in a H NMR spectrum (error
bars are omitted for clarity).
The sterically less hindered tert-butyl-substituted catalyst 4a
displays significantly lower activity than the sterically more hin-
dered triphenylsilyl-substituted catalyst 4b (Table 1, entry 6 vs.
7 and entry 12 vs. 13). This observation is in agreement to find-
ings of a previous study that compared the activity of com-
plexes 1a and 1b.[13a] The activity of complex 4a seems to be
comparable to complex 1a, whereas the activity of 4b falls be-
tween the phenoxytriamine complex 1b and the chiral com-
plex 2. Thus, 4b facilitates smooth cyclisation of gem-dialkyl
and gem-diphenyl substituted aminopentenes 5a–c as well as
substrates 5d and 5e containing a vinylic phenyl group at
room temperature (Table 1, entries 1, 7, 13, 18, and 23). The
cyclisation of the aminohexene derivative 5 f also proceeded
at room temperature, but required 3 days to reach completion
(Table 1, entry 28). Only the cyclisation of the N-benzylated
substrate 5g required heating of the reaction mixture (Table 1,
entry 33). Therefore, further kinetic and mechanistic studies
were performed with the more active and more stable catalyst
4b.
Figure 3. Logarithmic plot of normalized substrate concentration versus
time for the hydroamination/cyclisation of (1-allylcyclohexyl)methylamine
(5b, C0 =0.167 molLÀ1) in [D6]benzene at 258C using varying concentrations
of 4b.
Mechanistic and kinetic studies
A kinetic investigation utilizing complex 4b in the hydroamina-
tion/cyclisation of substrate 5b revealed a first-order rate de-
pendence on substrate concentration (Figure 2 and Figure 3),
as well as a first order rate dependence on catalyst concentra-
tion (Figure 4). The cyclisation of 5b with the sterically less de-
manding, tert-butyl-substituted catalyst 4a also proceeds with
a first-order rate dependence on substrate concentration (Fig-
ure S1 in the Supporting Information). These observations are
in agreement with the catalytic behaviour of complex 1a and
2; however, they contrast the behaviour of complex 1b, which
exhibits zeroth order rate dependence on substrate concentra-
tion.[13]
The cyclisation of the N-deutero substrate [D2]5b with com-
plex 4b exhibited a significant primary kinetic isotope effect
(KIE) of 3.9Æ0.2 (Figure 5), in agreement with previous obser-
vations for complexes 1b and 2.[13] Similar KIE’s have been also
Figure 4. Dependence of observed rate on catalyst concentration for the
hydroamination/cyclisation of (1-allylcyclohexyl)methylamine (5b,
C0 =0.167 molLÀ1) with catalyst 4b in [D6]benzene at 258C.
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7843
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
A kinetic study of the cyclisa-
tion of the less reactive gem-di-
methyl-substituted aminoalkene
5c allowed measuring reaction
rates in a wider temperature
range from 22 to 808C
(Figure 6). The activation param-
eters were determined as DH° =
10.5(Æ0.4) kcalmolÀ1 and DS° =
À45.7(Æ1.4) calKÀ1 molÀ1 from
the Eyring plot (Figure 7).
Table 1. Hydroamination of aminoalkenes catalysed by magnesium complexes 1a–c, 2, and 4a,b.[a]
Entry
Substrate
Product
Cat.
[mol%]
T
[8C]
t
[h]
Conv.
[%]
Ref.
1
2
3
4
5
4b (5)
1a (5)
1b (3)
1c (5)
2 (3)
22
25
25
25
22
3
4
3
5
99
99
99
98
99
this work
[13a]
[13a]
[13a]
[13b]
1.5
The phenoxyamine magnesi-
um catalyst 4b shows signs of
product inhibition, as indicated
by the observed rate depression
when substrate 5c is added in
two batches (Figure 8). This ap-
pears to be in contrast to obser-
vations made for complexes 1b
and 1c.[13a] However, this obser-
vation is in agreement with pre-
vious studies on binaphtholate
rare-earth metal complexes[4d]
6
7
8
9
10
11
4a (5)
4b (5)
1a (10)
1b (5)
1c (5)
2 (5)
60
22
25
25
25
22
4
4
12
13
22
2
98
98
98
96
81
99
this work
this work
[13a]
[13a]
[13a]
[13b]
12
13
14
15
16
17
4a (5)
4b (5)
1a (10)
1b (10)
1c (10)
2 (5)
110
22
100
100
100
22
48
36
40
18
40
10
78
96
98
98
48
97
this work
this work
[13a]
[13a]
[13a]
and
b-diketiminate
calcium
amide complexes.[12f]
The stoichiometric reaction of
4b with 2,2-dimethylpent-4-en-
1-amine (5c) in a 1:1 ratio did
not lead to any hydroamination
product formation within 5 h at
room temperature (Figure 9a).
Subsequent addition of a second
equivalent of 5c produced
1 equivalent of the hydroamina-
tion product 6c after 6 h at
room temperature (Figure 9b).
The sharpness of the signals
of 6c suggests that it is not
bound to the magnesium com-
plex, which is corroborated by
DFT findings that 6c binds less
strongly than 5c at magnesium
in the Mg amido catalyst com-
plex (vide infra, Figure 10). This
stoichiometric reaction supports
the hypothesis that the cyclisa-
tion proceeds only if more than
one equivalent of aminoalkene
substrate (with respect to the
magnesium catalyst) is present
in the reaction mixture. Hence, it
can be proposed that two sub-
strate molecules are bound to
the catalytic metal centre in
order for the reaction to pro-
[13b]
18
19
20
21
22
4b (5)
1a (10)
1b (10)
1c (10)
2 (5)
22
3
7
12
24
2
99
99
99
98
99
this work
[13a]
[13a]
[13a]
[13b]
100
100
100
22
23
24
25
26
27
4b (5)
1a (10)
1b (10)
1c (10)
2 (5)
22
25
25
25
22
2
2
1.5
2
0.1
99
99
99
99
99
this work
[13a]
[13a]
[13a]
[13b]
28
29
30
31
32
4b (5)
1a (10)
1b (10)
1c (10)
2 (5)
22
60
60
60
22
72
13
2.5
8
96
99
99
99
98
this work
[13a]
[13a]
[13a]
[13b]
6
33
4b (5)
80
48
95
this work
[a] Reaction conditions: 0.1 mmol substrate, 0.6 mL [D6]benzene, Ar atm.
observed for alkaline-earth metal,[12m–o,w] rare-earth met-
al[3c,w, z,4d] and Group 4 metal[3w,5s–v] catalysts.
ceed. This finding is reminiscent of tris(oxazolinyl)phenylborate
magnesium[12n] and cyclopentadienylphosphazene dialkyl rare-
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7844
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 5. First order plot for the cyclisation of (1-allylcyclohexyl)methylamine
(5b) and [D2](1-allylcyclohexyl)methylamine ([D2]5b, C0 =0.167 molLÀ1) in
the presence of 4b ([cat.]=8.40 mmolLÀ1) in [D6]benzene at 258C.
Figure 7. Eyring plot for the hydroamination/cyclisation of 2,2-dimethylpent-
4-en-1-amine (5c, C0 =0.167 molLÀ1) in [D6]benzene using 5.0 mol% of 4b
as catalyst.
Figure 8. Logarithmic plot of normalized substrate concentration versus
time for the hydroamination of 2,2-dimethylpent-4-en-1-amine (5c; first ad-
dition: C0 =0.167 molLÀ1) in [D6]benzene using 5.0 mol% of 4b
([cat.]=8.33 mmolLÀ1) as catalyst at 808C. Upon completion of the first run,
a second batch of substrate was added. The second run proceeds slower, in-
dicating product inhibition.
Figure 6. Logarithmic plot of normalized substrate concentration versus
time for the hydroamination of 2,2-dimethylpent-4-en-1-amine (5c,
C0 =0.167 molLÀ1) in [D6]benzene using 5.0 mol% of 4b as catalyst at vari-
ous temperatures.
earth metal[3z] catalysts that required the addition of more than
2 equiv of aminoalkene substrate before the hydroamination
commenced.
A difference between the concerted non-insertive mech-
anism and the two-step migratory insertion/protonolysis mech-
anism is the existence of
a metal-alkyl intermediate B
(Scheme 1). The presence of this intermediate for lanthano-
cene-catalysed hydroamination has been indicated by the suc-
cessful realization of hydroamination/carbocyclisation se-
quences.[3e,h,i] In an attempt to prove the intermediacy of spe-
cies B we investigated the reaction of N-allylpent-4-en-1-amine
(7) with catalyst 4b. However, the hydroamination product 8
was the only observed product (Scheme 5) and no bicyclic
product stemming from a hydroamination/carbocyclisation pro-
cess was detected. A similar observation was made in a study
Scheme 5. Attempted hydroamination/carbocyclisation of N-allylpent-4-en-
1-amine (7) with catalyst 4b.
by Hill and co-workers using nacnac calcium hydroamination
catalysts.[12f] However, the interpretation with respect to the in-
termediacy of species B in the catalytic process is not unambig-
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7845
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Computational studies
In order to increase our mecha-
nistic understanding we em-
barked on a detailed computa-
tional analysis of rival mechanis-
tic pathways for cyclisation of
the
gem-dimethyl-substituted
aminoalkene 5c (denoted there-
after as S) into pyrrolidine 6c
(denoted thereafter as P) medi-
ated by compound 4b (denoted
thereafter as C2). The DFT
method was employed as an es-
tablished and predictive means
to aid mechanistic understand-
ing;[18–20] the computational
methodology employed (disper-
sion-corrected B97-D3 in con-
junction with basis sets of triple-
z quality and a sound treatment
of bulk solvent effects, see the
Computational Details section
for details) simulated authentic
reaction conditions adequately
and mechanistic analysis is
based on Gibbs free-energy pro-
files. Furthermore, the validity of
the computational protocol em-
ployed for reliably mapping the
energy landscape of alkaline-
earth-mediated hydroamination
has been substantiated before[18i]
and this allowed mechanistic
conclusions with substantial pre-
dictive value to be drawn. Whilst
the presentation herein is re-
stricted exclusively on the most
accessible pathways, a complete
account of all scrutinised species
is given in the Supporting Infor-
mation.
Catalyst initiation
Effective hydroamination cataly-
sis entails the initial conversion
of starting material C2 (com-
pound 4b, Scheme 4) into the
catalytically competent phenoxy-
amine [{ONN}Mg(NHR)] magne-
sium amidoalkene compound
Figure 9. 1H NMR spectra of stoichiometric reaction of complex 4b and 2,2-dimethylpent-4-en-1-amine (5c) in
[D6]benzene at room temperature. a) Complex 4b/5c=1:1, no hydroamination product was observed after 5 h at
room temperature. b) Complex 4b/5c=1:2, 1 equivalent of hydroamination product 6c was observed after 6 h at
room temperature (§=toluene formed in reaction of 4b with 5c; P=hydroamination product 6c; *=residual
THF).
uous, because one can conclude only that either, 1) species B is
not an intermediate with a sufficient lifetime to allow the
carbocyclisation to proceed, or 2) the activation barrier for the
magnesium-catalysed carbocyclisation exceeds the barrier for
the other steps in the hydroamination process.
through protonolytic MgÀC alkyl bond cleavage. The initial
facile association[21] of one and two aminoalkene (S) molecules
is somewhat downhill in free energy, whilst association of
a third substrate molecule is seen to be thermodynamically un-
favourable. The MgÀC alkyl bond aminolysis commencing from
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7846
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 10. Amine association at the catalytically relevant [{ONN}Mg(NHR)] compound.[21,23]
substrate adducts C2·(S)n evolves through a metathesis-type
transition state (TS) that decays thereafter into substrate-free
or substrate-adducted forms of the magnesium amido catalyst
complex (C3·(S)n) through facile liberation of toluene. Pathways
with one or two substrate molecules involved appear equally
feasible.[22] However, the participation of one spectator sub-
strate molecule via C2+2S!C3·S stabilises both the TS struc-
ture and also the {ONN}Mg amido complex and thus favours
this pathway relative to C2+S!C3 on both thermodynamic
and kinetic grounds (Figure S2 in the Supporting Information).
The moderate activation barrier of 13.3 kcalmolÀ1 together
with a thermodynamic force of 11.8 kcalmolÀ1 characterises
the initial conversion of C2 into the magnesium amido catalyst
complex as a kinetically easy, essentially irreversible transfor-
mation.
ishes for mono-substrate-adducted species C3’·S and C3·S,
which are virtually identical energetically. It is worth noting
that species C3’’·S, featuring h1-amidoalkene/h2-aminoalkene
units are found less favourable thermodynamically than C3’·S
(Figure S3 in the Supporting Information). With a further sat-
uration of the coordinative sphere around magnesium by
a second associated amine molecule, the tendency to form
C3’·(S)2 decreases rapidly. All attempts to locate C3’·(S)2 with an
intact (k3-ONN)Mg ligation were unsuccessful and only species
featuring an (k2-ONN)Mg ligation with an substrate (NH)-sta-
bilised dissociated NMe2 arm could be located, which is more-
over placed well above C3·(S)2, with an intact (k3-ONN)Mg liga-
tion, in energy.
As revealed from Figure 10, DFT predicts that the catalytical-
ly competent [{ONN}Mg(NHR)] compound is predominantly
present as mono-substrate (C3·S, C3’·S) adducted species, to-
gether with a less populated bis-amine adduct C3·(S)2, all of
which are expected to participate in rapid amine association/
dissociation equilibria.[21]
The catalytically competent [{ONN}Mg(NHR)] compound
Amine substrate (S) or product (P) molecules can associate in
various ways with magnesium at the catalytically competent
{ONN}Mg amido compound C3 (or A in Scheme 1) to give rise
to a multitude of adduct species, all of which are expected to
participate in rapid association/dissociation equilibria.[21] The
relative stabilities of various located species with one or two
adducted amine molecules are collated in Figure 10. Different
isomers of a given species are denoted by subscript numbers
and also by a or b labelling (see below). A moderately encum-
bering phenoxyamine {ONN}Mg ligation favours amine associa-
tion at magnesium, which is capable of accommodating up to
two amine molecules. The association pattern for substrate-
free and substrate-adducted forms reflects the variation in spa-
tial demands around magnesium. For substrate-free forms,
species C3’ featuring an amidoalkene unit, which forms a weak
chelate by orienting its double bond proximal to magnesium,
is thermodynamically favourable relative to C3 with a mono-
hapto N-ligated amidoalkene unit. This energy gap almost van-
The s-insertive mechanism
Migratory olefin 1,2-insertion into the MgÀN amido s-bond
Several imaginable trajectories for insertive cyclisation featur-
ing an axial or equatorial approach of the olefin unit that start
from C3’ or its mono- (C3’·S) and bis-substrate (C3’·(S)2) ad-
ducted species have been examined. With regard to C3’·S and
C3’·(S)2 precursor species, two families of trajectories are con-
ceivable, which sees the MgÀN amido s-bond being trans
(C3a’·(S)n!C4a·(S)n, Scheme 6) or cis (C3b’·(S)n!C4b·(S)n,
Scheme 7) disposed relative to the phenoxy group.[22]
Common to all the trajectories studied is that NÀC bond for-
mation evolves through a four-centre TS structure describing
a metal-mediated migratory olefin insertion into the MgÀN
amido s-bond, which occurs at distances of approximately 2 Š
for the emerging NÀC bond. Following the reaction path fur-
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7847
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 6. Alternative trajectories for migratory olefin insertion into the MgÀN amido s-bond at C3’ (top) and its substrate adducted species C3a’·S (middle)
and C3a’·(S)2 (bottom), to involve substrate adducted species where amido is trans to the phenoxy group.
Scheme 7. Alternative trajectories for migratory olefin insertion into the MgÀN amido s-bond at substrate adducted species C3b’·S (top) and C3b’·(S)2
(bottom), to involve substrate adducted species where amido is cis to the phenoxy group.
ther, TS structures decay into {ONN}Mg alkyl intermediate C4
(or B in Scheme 1), which has the azacycle bound to magne-
sium through its methylene tether and the N donor centre.
The condensed energy profiles in Figure 11 (exemplified for
substrate adducted species featuring amido cis to the phenoxy
group) reveal that substrate-free and mono-substrate-adduct-
ed {ONN}Mg species C3’ and C3’·S are similarly competent at
effecting insertive cyclisation, whilst enhanced steric pressure
around the Mg centre through association of another substrate
molecule renders migratory olefin insertion at C3’·(S)2 substan-
tially less favourable on both thermodynamic and kinetic
grounds. Increased steric pressure along C3’·(S)2!C4·(S)2 path-
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7848
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ways is a direct consequence of
the necessity of activating the
olefin unit towards NÀC bond
formation by close contact with
the electropositive metal centre
and is manifested in the fact
that species participating via
C3’·(S)2!C4·(S)2 all feature a {k2-
ONN}Mg ligation with an sub-
strate(NH)-stabilised dissociated
NMe2 arm. Irrespective of wheth-
er one or two spectator sub-
strate molecules are involved or
whether the MgÀN amido s-
bond is trans (C3a’(S)n!C4a(S)n)
or cis (C3b’(S)n!C4b(S)n) dis-
posed relative to the phenoxy
group, an equatorial approach
of the olefin unit onto the MgÀN
bond is found most favour-
able.[22]
Migratory olefin insertion fol-
lowing an equatorial trajectory
preferably proceeds into the
MgÀN amido s-bond that is
trans to the phenoxy unit and
favours C3a’·S!C4a·S slightly
over C3’!C4. The energetically
Figure 11. Most accessible pathways for migratory olefin insertion into the Mg MgÀN amido s-bond at C3’ and its
substrate adducted species C3b’·S and C3b’·(S)2 to involve substrate adducted species featuring amido cis to the
phenoxy unit.[23]
prevalent C3a’·S!C4a·S insertive cyclisation has a modest
barrier of 15.1 kcalmolÀ1 (relative to the prevalent C3a1’·S form
of the {ONN}Mg amido complex) to overcome and is ender-
gonic (Figure S4a4 in the Supporting Information). For precur-
sor species C3b’·(S)n, which have amido and phenoxy groups
cis disposed, olefin insertion is likely proceeding through
C3b’·S!C4b·S, but is slightly more demanding kinetically
(DG° =16.8 kcalmolÀ1, Figure 11, relative to the prevalent
C3a1’·S form of the {ONN}Mg amido complex). Hence, both
C3a’·S!C4a·S and C3b’·S!C4b·S are likely traversed as most
accessible pathways, although the latter with a somewhat
smaller probability. All this characterises insertive ring closure
for a sterically moderately encumbered {ONN}Mg amido cata-
lyst species as a kinetically facile, reversible step that favours
C3’·S. We note that a previous computational study[20] on cy-
clohydroamination of aminoalkene S by a tris(oxazolinyl)borate
magnesium amido catalyst also revealed a kinetically modest
and reversible insertion step with a barrier of similar magni-
tude (DG° =16.8 kcalmolÀ1),[20] but without the involvement of
spectator amine molecules.
C3b·(S)n!C4b·(S)n (Scheme 9) pathways for insertive cyclisa-
tion.[22]
Aminolysis evolves through a metathesis-type TS structure
that describes the cleavage of an already suitably polarised NÀ
H bond and concurrent CÀH bond formation. The process to
commence at C4a·S, C4b·S involves species with an intact {k3-
ONN}Mg ligation. In contrast, for intramolecular H transfer to
occur at bis-amine adducts C4a·(S)2, C4b·(S)2, featuring an sub-
strate NÀH stabilised dissociated NMe2 arm, a {k2-ONN}Mg liga-
tion is maintained throughout. It places pathways with two
substrate molecules participating at a higher energy than path-
ways with a single substrate molecule involved (Figure 12).
Hence, excess substrate is unlikely facilitating MgÀC bond
aminolysis on either thermodynamic or kinetic grounds.
Interestingly, C4a·S, C4b·S generated via preferably tra-
versed pathways for insertive cyclisation to follow an equator-
ial trajectory are precursor species for the most accessible
C4a·S!C5 and C4b·S!C5 pathways. Thus, neither association
of another substrate molecule or substantial relaxation within
the ligand sphere is required for protonolytic MgÀC bond
cleavage to occur at the {ONN}Mg alkyl intermediate. The two
C4a·S!C5 and C4b·S!C5 pathways have an identical intrin-
sic barrier (relative to its direct precursors C4a·S, C4b·S, re-
spectively), but the stability gap between C4a·S and C4b·S, in
favour of the latter, renders C4b·S!C5 the most accessible
pathway for MgÀC alkyl bond aminolysis (Figure 12, Fig-
ure S5a1 and Figure S5a3 in the Supporting Information). It
features an intrinsic barrier of 11.9 kcalmolÀ1 (Figure 12, rela-
tive to C4b·S) and affords [{ONN}Mg(NHR)·(NHRR)] compound
MgÀC azacycle tether aminolysis
The {ONN}Mg alkyl intermediate formed via the most accessi-
ble C3a’·S!C4a·S and C3b’·S!C4b·S pathways already has
one substrate molecule complexed at magnesium, which is
moreover favourably placed cis to the MgÀC linkage. Several
trajectories for intramolecular H transfer have been probed for
species generated via C3a·(S)n!C4a·(S)n (Scheme 8) and
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7849
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 8. Alternative trajectories for MgÀC azacycle tether aminolysis at the magnesium alkyl intermediate, which is generated via C3a·(S)n!C4a·(S)n,
through pathways with one (top) or two (bottom) adducted substrate molecules involved.
Scheme 9. Alternative trajectories for MgÀC azacycle tether aminolysis at the magnesium alkyl intermediate, which is generated via C3b·(S)n!C4b·(S)n,
through pathways with one (top) or two (bottom) adducted substrate molecules involved.
C5 in a process that is downhill by 8.3 kcalmolÀ1 (relative to
C4b·S). The initially formed isomer C51 with an axially bound
amido group is readily transformed thereafter into a thermo-
dynamically more favourable isomer C53 featuring an equator-
ial amido.[22] The facile displacement of pyrrolidine P by sub-
strate S at C5 renders aminolysis even more favourable
thermodynamically. The DFT predicted gap of 3.6 kcalmolÀ1
(DDG°, Figure 12, Figure S5a1 and Figure S5a3 in the Support-
ing Information) between most accessible trajectories via
C4a·S!C5 and C4b·S!C5 makes the former almost impos-
sible to be traversed.
sis step via the most accessible C4b·S!C5 pathway. Participa-
tion of another coordinated amine molecule, however, is un-
likely to facilitate the process on either thermodynamic or kin-
etic grounds. It is worth nothing that the effective inaccess-
ibility of rival C4a·S!C5 pathway has its primary origin in the
stability gap between C4a·S and C4b·S, but not in their intrin-
sic reactivity, which is of comparable magnitude.
Generation of a {ONN}Mg pyrrolide compound
A magnesium pyrrolide C6·(S)n may represent a relevant off-
cycle intermediate and could thus play a role in hydroamina-
tion catalysis by the system at hands. Provided that the pyrro-
lide is kinetically accessible at affordable costs, it could well
compete with C3·(S)n and C5·(S)n for representing the catalyst
Overall, the DFT-derived smooth energy profile shown in
Figure 12 is indicative of a kinetically affordable and exergonic,
hence (together with subsequent downhill product displace-
ment) irreversible, intramolecular MgÀC alkyl bond protonoly-
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7850
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
torial pyrrolide group is the
prevalent form of the {ONN}Mg
pyrrolide compound and bis-
substrate adducted C6·(S)2 and
substrate-free C6 species are at
higher energy.[22]
The concerted proton-triggered
cyclisation mechanism
The
alternative
mechanistic
pathway outlined in Scheme 2,
which describes NÀC ring clos-
ure triggered by concomitant
amino proton transfer onto the
adjacent C=C linkage, thereby
affording cycloamine products
through a concerted single-step
transformation, is studied next.
Figure 12. Most accessible pathways for MgÀC s-bond aminolysis at amine-adducted species C4b·S and C4b·(S)2
of the {ONN}Mg alkyl intermediate that are formed via C3b·(S)n!C4b·(S)n for insertive cyclisation.[23]
resting state. Noteworthy, a mag-
nesium pyrrolide compound has
been observed by NMR spectros-
copy upon monitoring aminoal-
kene cyclisation by a tris(oxazoli-
nyl)borate {ToM}MgII catalyst sys-
tem.[12n]
To this end, plausible routes
for generation of a {ONN}Mg
pyrrolide have been examined,
including: 1) intramolecular 1,3
H transfer at the Mg alkyl inter-
mediate C4·(S)n, a process that
may benefit from participation
of excess amine through a relay-
type mechanism, and 2) cyclo-
amine a-proton abstraction at
intermediate C5·(S)n. Similar to
Figure 13. Most accessible pathways for conversion of {ONN}Mg amido cycloamine compound C5 into {ONN}Mg
pyrrolide intermediate C6.[23]
our findings for
{ToM}MgII catalyst,[20] route
proves to be far more accessible
a
related
B
when compared to route A and is likely the dominant path-
ways that leads to C6·(S)n.
The two trajectories examined involving NÀC bond formation
through an axial olefin approach at the MgÀN amido s-bond,
triggered by a proton transferred from an equatorially bound
substrate, or alternatively describing concurrent ring closure at
an equatorial MgÀN bond and proton delivery from an axially
bound S (Scheme 10), are found to be equally viable, with the
trajectory for axial olefin approach being somewhat pre-
ferred.[22]
The energy profiles of the most accessible pathways to com-
mence from C5 and its mono-substrate adducted form, shown
in Figure 13, reveal that a-proton abstraction preferably pro-
ceeds at C5, which is initially formed via the dominant
C4b·S!C5 pathway for protonolytic MgÀC bond cleavage.
The process is unlikely benefiting from complexation of an-
other substrate molecule, thus paralleling the findings for the
protonolysis step. Proton transfer evolves through a metathe-
sis-type TS[C5–C6·S] with rather modest intrinsic barriers of 4.8
(C5!C6·S) and 15.5 kcalmolÀ1 (C6·S!C5) to afford substrate
adducted {ONN}Mg pyrrolide C6·S in a process that is driven
by a thermodynamic force of 10.7 kcalmolÀ1 (relative to C5).
Thus, C5QC6·S conversion is kinetically feasible and reversible,
but strongly favours C6·S. DFT predicts that C6·S with an equa-
The most accessible pathway following this trajectory starts
with a facile conversion of the thermodynamically prevalent
C3a’·S into a conformer that has olefin and substrate moieties
already favourably arranged and evolves through a six-centre
TS structure that constitutes NÀC5 ring closure with concomi-
tant transfer of a proton from a magnesium-bound and,
hence, acidified substrate molecule on the adjacent olefin-C6
centre. As exemplified in Figure 14 for the more accessible tra-
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7851
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 10. Alternative trajectories for concerted proton-triggered NÀC ring
closure at C3·S (top) and C3·(S)2 (bottom) adducts of the [{ONN}Mg(NHR)]
complex.
Figure 14. Selected structural parameter [Š] of the optimised TS structure
for proton-assisted concerted NÀC/CÀH bond formation at C3a’·S via a tra-
jectory for axial olefin approach.[24]
jectory of axial olefin approach,
the located TS[C3a’·S–C5] fea-
tures a {k3-ONN}Mg ligation with
a NÀC5 distance of 1.92 Š and
vanishing NÀH (1.17 Š) and
emerging CÀH (1.65 Š) bonds,
which is indicative of a concert-
ed, but asynchronous proton
transfer. The rather large MgÀ
olefin distance (>3.6 Š) de-
scribes NÀC ring closure occur-
ring outside of the immediate
vicinity of the alkaline earth,
thereby reflecting that amino
proton delivery activates the C=
C linkage and not the close in-
teraction with the electropositive
metal as necessitated by the
rival s-insertive pathway. After
Figure 15. Most accessible pathways for NÀC bond formation with concurrent delivery of the amino proton to
the olefin unit at the [{ONN}Mg(NHR)] compound with one (C3·S, top) or two (C3·(S)2, bottom) adducted substrate
molecules.[23]
passing through the TS struc-
ture,
TS[C3a’·S–C5]
and
TS[C3·(S)2–C5a·S] decay into the
{ONN}Mg amino cycloamine
compound C5 and its substrate adduct C5a·S, respectively.
The dominant pathway for concurrent NÀC ring closure and
amino proton delivery at axial sites in C3a’·S has a barrier of
23.1 kcalmolÀ1 (relative to the prevalent C3a1’·S form of the
{ONN}Mg amido complex) to overcome and affords C5 in a pro-
cess that is downhill by À6.3 kcalmolÀ1 (Figure 15). Thus,
proton-assisted concerted cyclisation is irreversible as is MgÀC
alkyl bond aminolysis and both steps furnish compound C5.
Similar to previous discussion for the aminolysis step, the asso-
ciation of another spectator substrate molecule does not serve
stabilising product or TS structures and additional amine mole-
cules are therefore unlikely to participate in this step
(Figure 15). Noteworthy, TS[C3·(S)2–C5a·S] still features an
intact {k3-ONN}Mg ligation whilst for protonolysis the participa-
tion of an additional spectator substrate molecule provokes
dissociation of an NMe2 arm. This behaviour reflects the differ-
ences in spatial demands imposed at the metal centre whilst
traversing s-insertive and concerted proton-triggered mecha-
nistic pathways.
Comparison of mechanistic pathways
Figure 16 collates the free-energy profiles assessed for alterna-
tive mechanistic scenarios for cyclohydroamination of gem-di-
methyl-substituted aminoalkene S (5c) to be mediated by
phenoxyamine {ONN}Mg alkyl starting material 4b (R=SiPh3).
It focuses exclusively on the most accessible pathways to com-
mence from the catalytically competent phenoxyamine
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7852
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 16. Condensed reaction profile for intramolecular hydroamination of gem-dimethyl-substituted aminoalkene 5c (S) to be mediated by a phenoxyamine
{ONN}Mg alkyl compound 4b (R=SiPh3) proceeding through alternative mechanistic cycles. Cycloamine product (P) liberation through C5+S!C3a’·S+P is
included.[25]
[{ONN}Mg(NHR)] amido compound C3, which is readily gener-
ated through facile initial catalyst activation, whereas other
less probable or energetically prohibitive pathways have been
omitted. It is worth noting that the thermodynamic force for
S!P cyclisation has been subtracted from C5, C6·S and
TS[C5–C6·S].
It becomes clear from Figure 16 that substrate adducted
{ONN}Mg pyrrolide C6·S is the most stable species associated
to the catalytic cycle and thus likely corresponds to the cata-
lyst resting state. It is readily converted into magnesium amido
cycloamine compound C5; its most stable conformer C53 is
7.1 kcalmolÀ1 higher in energy than C6·S. Subsequent displace-
ment of cycloamine by amine substrate gives rise to the
{ONN}Mg amido complex in a supposedly kinetically easy,
downhill transformation. DFT predicts that the magnesium
amido active catalyst complex C3 is predominantly present as
mono-substrate (C3·S, C3’·S) adducted species, the various pos-
sible conformers of which are expected to participate in rapid
equilibria.[21]
s-insertive pathways shown in Figure 16 is equally consistent.
This pathway entails: 1) facile and reversible s-insertive NÀC
bond forming cyclisation via C3a’·S!C4a·S or C3b’·S!C4b·S
that favours the Mg amido catalyst species; 2) downhill turn-
over-limiting C4b·S!C5 Mg–C s-bond aminolysis at the
{ONN}Mg alkyl intermediate; followed by 3) kinetically easy
and exergonic displacement of cycloamine by substrate to re-
generate the {ONN}Mg amido active catalyst complex.
The two mechanistic cycles account equally well for all the
observed features of the process, are indistinguishable by an
empirical rate law and are driven by a thermodynamic force of
identical magnitude. DFT reveals that a proton-assisted con-
certed NÀC/CÀH bond forming pathway is energetically pro-
hibitive in the presence of a kinetically less demanding s-inser-
tive pathway. The magnitude of the assessed kinetic disparity
(DDG° =5.6 kcalmolÀ1, Figure 16) is indicative that the s-inser-
tive pathway is likely traversed exclusively. The DFT predicted
total barrier of 23.1 kcalmolÀ1 (relative to the catalyst resting
state C6·S) for magnesiumÀalkyl bond aminolysis matches the
experimentally determined Eyring parameter (DG° =24.1(Æ
0.6) kcalmolÀ1 (298 K)) gratifyingly well. DFT estimates a classical
KIE of 2.7 (298 K) to be associated with turnover-limiting
aminolysis via TS[C4b3·S–C51].[32] Taking tunnelling into ac-
count,[32b] a primary KIE of 3.1 (298 K) was obtained, which is
reasonably close[32d] to the observed value of 3.9 (vide supra).
The NÀC ring closure with concurrent amino proton transfer
taking place outside the immediate proximity of the electro-
positive magnesium centre via a six-centre TS[C3a’·S–C5] is (in
combination with ensuing facile cycloamine displacement) irre-
versible[25a] and has a total barrier of 28.7 kcalmolÀ1 (relative to
the catalyst resting state C6·S) to overcome. The observed
large primary KIE is easily explained by this pathway, but the
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7853
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
thyl)amino)methyl) (3a)[28] and the aminoalkene substrates 2,2-di-
phenylpent-4-enyl amine (5a),[3k] (1-allylcyclohexyl)methylamine
(5b),[8g] 2,2-dimethyl-pent-4-enyl amine (5c),[29] (E)-2,2-dimethyl-5-
phenylpent-4-en-1-amine (5d),[4d] 2,2,5-triphenylpent-4-enyl amine
(5e),[30] 2,2-diphenylhex-5-enyl amine (5 f),[31] 1-(1-allylcyclohexyl)-N-
benzylmethanamine (5g)[8g] and N-allylpent-4-en-1-amine (7)[3e]
were synthesized according to literature protocols. The hydro-
amination products 2-methyl-4,4-diphenylpyrrolidine (6a),[3k] 3-
methyl-2-azaspiro[4,5]decane (6b),[3o] 2,4,4-trimethylpyrrolidine
(6c),[3c] 2-benzyl-4,4-dimethylpyrrolidine (6d),[4d] 2-benzyl-4,4-
diphenylpyrrolidine (6e),[5i] 2-methyl-5,5-diphenylpiperidine (6 f),[3m]
2-benzyl-3-methyl-2-azaspiro[4.5]decane (6g),[8g] and 1-allyl-2-
methylpyrrolidine (8)[12f] are known compounds and were identi-
fied by comparison to the literature NMR spectroscopic data.
Conclusion
In the quest to develop the mechanistic insight into the intra-
molecular hydroamination of aminoalkenes utilizing novel
phenoxyamine magnesium compounds, a complementary syn-
thetic and computational study has been conducted. For the
well characterised cyclisation of gem-dimethyl-substituted
aminoalkene 5c mediated by a phenoxyamine magnesium
alkyl starting material 4b, a reliable DFT protocol (dispersion-
corrected B97-D3 in conjunction with basis sets of triple-z
quality and a sound treatment of bulk solvent effects) has
been employed as an established and predictive means to
scrutinise rival mechanistic pathways. Two plausible mechanis-
tic scenarios are compatible with observed distinct process fea-
tures, which include substantial primary KIEs and a second-
order rate law. Firstly, a proton-triggered concerted NÀC/CÀH
bond-forming mechanism, thereby accomplishing amidoal-
kene!cycloamine conversion in a single step and secondly,
a stepwise s-insertive pathway that involves rapid and revers-
ible migratory olefin insertion into the MgÀN amido bond
linked to irreversible and slower MgÀC alkyl bond aminolysis.
The comprehensive computational examination discloses non-
competitive kinetic demands for a pathway describing NÀC
ring closure taking place outside of the immediate proximity
of the electropositive metal centre triggered by concurrent
amino proton delivery at the C=C linkage evolving through
a six-centre TS structure. It, thus, militates against the opera-
tion of a concerted non-insertive cyclisation pathway in the
presence of a kinetically less demanding s-insertive pathway.
The magnitude of the assessed kinetic disparity indicated that
the s-insertive pathway is likely traversed exclusively. This
pathway entails: 1) facile and reversible s-insertive NÀC bond
forming cyclisation via C3a’·S!C4a·S or C3b’·S!C4b·S that
favours the Mg amido catalyst species; 2) downhill turnover-
limiting C4b·S!C5 MgÀC s-bond aminolysis at the {ONN}Mg
alkyl intermediate; followed by, 3) kinetically easy and exergon-
ic displacement of cycloamine by substrate to regenerate the
{ONN}Mg amido active catalyst complex. The DFT predicted ef-
fective barrier (relative to the {ONN}Mg pyrrolide resting state)
for magnesiumÀalkyl bond aminolysis matches the empirically
determined Eyring parameters gratifyingly well.
Synthesis and characterisation
Synthesis of 4-(tert-butyl)-2-(((2-(dimethylamino)ethyl)(methyl)-
amino)methyl)-6-(triphenylsilyl)phenol (3b): 5-(tert-Butyl)-2-hy-
droxy-3-(triphenylsilyl)benzaldehyde (872 mg, 2.0 mmol)[27] and
N,N,N’-trimethylethylenediamine (612 mg, 6.0 mmol) were mixed in
1,2-dichloroethane (10 mL) and then treated with sodium triacet-
oxyborohydride (1.3 g, 6.0 mmol) and AcOH (360 mg, 6.0 mmol).
The mixture was stirred at room temperature under a nitrogen at-
mosphere for 2 days. The reaction mixture was quenched by addi-
tion of aqueous saturated NaHCO3 (100 mL), and the product was
extracted with diethyl ether (3”50 mL). The combined organic
layers were washed with water (3”100 mL) and dried over MgSO4.
The solvent was evaporated under vacuum to give the product as
a white foam which was used without further purification. Yield:
845 mg (81%). 1H NMR (500 MHz, CDCl3): d=7.63 (m, 6H, Si(C6H5)3),
7.36 (m, 9H, Si(C6H5)3), 7.05 (m, 2H, aryl-H), 3.70 (s, 2H, ArCH2N),
2.55 (m, 2H, NMeCH2CH2NMe2), 2.38 (m, 2H, NMeCH2CH2NMe2),
2.31 (s, 3H, NCH3), 2.14 (s, 6H, N(CH3)2), 1.11 ppm (s, 9H, C(CH3)3);
13C{1H} NMR (125 MHz, CDCl3): d=161.1, 141.0, 136.6, 135.8,
134.7, 129.2, 128.4, 127.7, 120.9, 119.4 (aryl), 61.7 (ArCH2N),
57.1
(NMeCH2CH2NMe2),
54.4
(NMeCH2CH2NMe2),
45.6
(NMeCH2CH2N(CH3)2), 42.3 (N(CH3)CH2CH2NMe2), 34.1 (C(CH3)3),
31.7 ppm (C(CH3)3).
[(3a)Mg(CH2Ph)] (4a): In the glovebox, an NMR tube was charged
with 3a (16.0 mg, 0.05 mmol), Mg(CH2Ph)2(THF)2 (17.5 mg,
0.05 mmol) and 0.6 mL of [D6]benzene. Attempts to isolate com-
plex 4a have failed so far, due to facile ligand redistribution pro-
cesses during removal of solvent. Hence, complex 4a was pre-
1
pared and used in situ for catalytic experiments. H NMR (400 MHz,
[D6]benzene): d=7.62 (d, 4JH,H =3.0 Hz, 1H, aryl-H), 7.10 (m, 4H,
Taking all these aspects together, the mechanistic analysis
by complementary experimental and computational ap-
proaches presented herein provides compelling evidence that
aminoalkene hydroamination proceeds through a stepwise s-
insertive pathway in the presence of a catalytically competent
phenoxyamine magnesium amido compound.
4
CH2C6H5), 6.91 (d, JH,H =3.0 Hz, 1H, aryl-H), 6.77 (m, 1H, CH2C6H5),
3.20 (d, 2JH,H =12.0 Hz, 1H, ArCH2N), 3.10 (d, 2JH,H =12.0 Hz, 1H,
ArCH2N), 2.07 (m, 1H, CH2CH2), 1.79 (s, 9H, C(CH3)3, 2H, CH2C6H5),
1.74 (m, 3H, NCH3, m, 1H, CH2CH2), 1.48 ppm (s, 9H, C(CH3)3).
[(3b)Mg(CH2Ph)] (4b): 3b (104 mg, 0.2 mmol) was dissolved in
benzene (0.5 mL) and then Mg(CH2Ph)2(THF)2 (70 mg, 0.2 mmol)
was added. The reaction mixture was kept at room temperature
for 5 min and then the solvent was removed under vacuum. The
white residue was washed with hexanes (2”2 mL) at À508C and
the product was collected and dried in vacuo for 1 h. Yield:
102 mg (80%). 1H NMR (400 MHz, [D6]benzene): d=8.02 (m, 6H,
Experimental Section
General considerations
4
Si(C6H5)3), 7.82 (d, JH,H =2.8 Hz, 1H, aryl-H), 7.20 (m, 9H, Si(C6H5)3,
2
All operations were performed under an inert atmosphere of nitro-
gen or argon using standard Schlenk-line or glovebox techniques.
Solvents and reagents were purified as stated previously.[13]
Mg(CH2Ph)2(THF)2
benzaldehyde,[27] 2,4-di-tert-butyl-6-(((2-(dimethylamino)ethyl)(me-
4H, CH2C6H5, 1H aryl-H), 6.82 (m, 1H, aryl-H), 3.62 (d, JH,H =12.8 Hz,
1H, ArCH2N), 2.55 (d, 2JH,H =12.8 Hz, 1H, ArCH2N), 2.24 (m, 1H,
CH2CH2), 1.71 (s, 2H, CH2C6H5), 1.62 (s, 3H, NCH3), 1.52 (m, 1H,
CH2CH2), 1.34 (s, 9H, C(CH3)3), 1.30 (br, 6H, N(CH3)2), 0.93 ppm (m,
2H, CH2CH2); 13C{1H} NMR (125 MHz, [D6]benzene): d=171.3, 156.3,
[26]
and 5-(tert-butyl)-2-hydroxy-3-(triphenylsilyl)-
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org 7854
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
137.9, 137.8, 137.0, 136.2, 131.0, 129.3, 129.0, 128.5, 124.8, 122.5,
121.4, 117.9 (aryl), 63.7 (ArCH2N), 56.6 (NMeCH2CH2), 50.4
(NMeCH2CH2), 43.6 (NCH3), 34.4 (C(CH3)3), 32.5 (C(CH3)3), 20.5 ppm
(CH2C6H5).
for experimental condensed phase conditions. Calculated struc-
tures were visualised by employing the StrukEd program,[43] which
was also used for the preparation of 3D molecule drawings.
Acknowledgements
General procedure for NMR-scale catalytic hydroamination/
cyclisation reactions
Financial support by the ACS Petroleum Research Fund (PRF
9109-ND1) is gratefully acknowledged.
In the glovebox, a screw-cap NMR tube was charged with ferro-
cene (6.0 mg, 32.3 mmol), the catalyst (5.0 mmol) [D6]benzene
(0.6 mL), and the substrate (0.10 mmol). The NMR tube was sealed,
removed from the glovebox and placed in the thermostated oil
Keywords: alkaline-earth metals
·
density functional
1
bath or thermostated H NMR probe (Æ0.58C) and the conversion
calculations
mechanisms
·
hydroamination
·
magnesium reaction
·
was monitored by NMR spectroscopy by following the disappear-
ance of the olefinic signals of the substrate relative to the internal
standard ferrocene. NMR spectra were taken in appropriate time
intervals (1.5 or 3 min). The error of measurement (shown as error
bars in the raw kinetic plot) was estimated to Æ5% based on the
accuracy of integrating signals in a 1H NMR spectrum. Errors on
linear correlations were estimated from the standard error of the
linear regression analysis performed using Microsoft Excel.
[1] a) A. Ricci, Modern Amination Methods, Wiley-VCH, Weinheim, 2000;
b) A. Ricci, Amino Group Chemistry: From Synthesis to the Life Sciences,
Wiley-VCH, Weinheim, 2008.
[2] For reviews on hydroamination see: a) R. Taube, in Applied Homogene-
ous Catalysis (Eds.: B. Cornils, W. A. Herrmann), Wiley-VCH, Weinheim,
1996, Vol. 1, p.507; b) T. E. Müller, M. Beller, Chem. Rev. 1998, 98, 675;
c) J. J. Brunet, D. Neibecker, in Catalytic Heterofunctionalization from Hy-
droamination to Hydrozirconation (Eds.: A. Togni, H. Grützmacher),
Wiley-VCH, Weinheim, 2001, p.91; d) T. E. Müller, K. C. Hultzsch, M. Yus,
F. Foubelo, M. Tada, Chem. Rev. 2008, 108, 3795; e) S. Doye, in Science of
Synthesis (Ed: D. Enders), Thieme, Stuttgart, 2009, Vol. 40a, p.241;
f) A. L. Reznichenko, K. C. Hultzsch, Top. Organomet. Chem. 2013, 43, 51;
g) N. Nishina, Y. Yamamoto, Top. Organomet. Chem. 2013, 43, 115.
[3] Reviews on rare-earth metal-based catalysts: a) S. Hong, T. J. Marks, Acc.
Chem. Res. 2004, 37, 673; b) A. L. Reznichenko, K. C. Hultzsch, Struct.
Bonding (Berlin) 2010, 137, 1. For selected example of rare-earth metal-
catalysed hydroaminations of aminoalkenes see: c) M. R. Gagne, C. L.
Stern, T. J. Marks, J. Am. Chem. Soc. 1992, 114, 275; d) M. A. Giardello,
V. P. Conticello, L. Brard, M. R. GagnØ, T. J. Marks, J. Am. Chem. Soc. 1994,
116, 10241; e) Y. Li, T. J. Marks, J. Am. Chem. Soc. 1998, 120, 1757; f) Y. K.
Kim, T. Livinghouse, J. E. Bercaw, Tetrahedron Lett. 2001, 42, 2933;
g) Y. K. Kim, T. Livinghouse, Angew. Chem. Int. Ed. 2002, 41, 3645; Angew.
Chem. 2002, 114, 3797; h) G. A. Molander, S. K. Pack, Tetrahedron 2003,
59, 10581; i) G. A. Molander, S. K. Pack, J. Org. Chem. 2003, 68, 9214;
j) P. N. O’Shaughnessy, P. D. Knight, C. Morton, K. M. Gillespie, P. Scott,
Chem. Commun. 2003, 1770; k) S. Hong, S. Tian, M. V. Metz, T. J. Marks,
J. Am. Chem. Soc. 2003, 125, 14768; l) J.-S. Ryu, T. J. Marks, F. E. McDo-
nald, J. Org. Chem. 2004, 69, 1038; m) F. Lauterwasser, P. G. Hayes, S.
Bräse, W. E. Piers, L. L. Schafer, Organometallics 2004, 23, 2234; n) K. C.
Hultzsch, F. Hampel, T. Wagner, Organometallics 2004, 23, 2601; o) J. Y.
Kim, T. Livinghouse, Org. Lett. 2005, 7, 1737; p) J. Collin, J.-D. Daran, O.
Jacquet, E. Schulz, A. Trifonov, Chem. Eur. J. 2005, 11, 3455; q) E. Lu, W.
Gan, Y. Chen, Organometallics 2009, 28, 2318; r) H. F. Yuen, T. J. Marks,
Organometallics 2009, 28, 2423; s) T. K. Panda, C. G. Hrib, P. G. Jones, J.
Jenter, P. W. Roesky, M. Tamm, Eur. J. Inorg. Chem. 2008, 4270; t) L. J. E.
Stanlake, L. L. Schafer, Organometallics 2009, 28, 3990; u) I. Aillaud, D.
Lyubov, J. Collin, R. Guillot, J. Hannedouche, E. Schulz, A. Trifonov, Orga-
nometallics 2008, 27, 5929; v) D. V. Vitanova, F. Hampel, K. C. Hultzsch, J.
Organomet. Chem. 2011, 696, 321; w) K. Manna, M. L. Kruse, A. D.
Sadow, ACS Catal. 2011, 1, 1637; x) C. Queffelec, F. Boeda, A. PouilhØs,
A. Meddour, C. Kouklovsky, J. Hannedouche, J. Collin, E. Schulz, Chem-
CatChem 2011, 3, 122; y) Y. Chapurina, H. Ibrahim, R. Guillot, E. Kolod-
ziej, J. Collin, A. Trifonov, E. Schulz, J. Hannedouche, J. Org. Chem. 2011,
76, 10163; z) N. K. Hangaly, A. R. Petrov, K. A. Rufanov, K. Harms, M. Elf-
ferding, J. Sundermeyer, Organometallics 2011, 30, 4544; aa) P. Benn-
dorf, J. Kratsch, L. Hartenstein, C. M. Preuss, P. W. Roesky, Chem. Eur. J.
2012, 18, 14454.
Computational details
All calculations based on Kohn–Sham density functional theory
(DFT)[33] were performed by means of the program package TUR-
BOMOLE[34] using the B97-D[35] generalized gradient approximation
(GGA) functional within the RI-J integral approximation[36] in con-
junction with flexible basis sets of triple-z quality. Empirical disper-
sion corrections by Grimme (D3 with Becke–Johnson damping)[37]
were used to account for critical non-covalent interactions involved
in the studied HA catalysis. For magnesium we used the
(14s9p3d)/[5s5p3d] (def2-TZVPPD) all-electron basis set.[38] All re-
maining elements were represented by Ahlrich’s valence triple-z
def2-TZVPD basis set[38,39a,b] with polarization functions on all
atoms. The validity of the computational protocol employed for re-
liably mapping the energy landscape of alkaline-earth-mediated
hydroamination has been substantiated before[18i] and this allowed
mechanistic conclusions with substantial predictive value to be
drawn.
The free-energy landscape of the entire HA course was assessed
for gem-dimethyl-substituted aminoalkene 5c together with the
phenoxyamine {ONN}Mg alkyl complex 4b (R=SiPh3) as starting
material. No structural simplification of any of the key species in-
volved was imposed. The DFT calculations have simulated the au-
thentic reaction conditions by treating the bulk effects of the ben-
zene solvent by a consistent continuum model in form of the con-
ductor-like screening model for realistic solvents (COSMO-RS).[40]
This solvation model includes continuum electrostatic and also
solvent-cavitation and solute-solvent dispersion effects through
surface-proportional terms. The free solvation enthalpy has been
assessed with the aid of COSMO-RS at the BP86[41]/(def2-TZVPPD+
def2-TZVPD)//BP86-D3/(def2-TZVPPD+def2-TZVPD)[38] level of ap-
proximation. Geometry optimisation and frequency calculations
were performed at the B97-D3/(def2-TZVPPD+SV(P))[39c] level to
confirm the nature of all optimised key structures and to deter-
mine thermodynamic parameters (298 K, 1 atm) under the conven-
tional ideal-gas, rigid-rotor and quantum-mechanical harmonic-os-
cillator approximation. The entropy contributions for condensed-
phase conditions were estimated based on computed gas-phase
entropies by employing the procedure of Okuno.[42] The mechanis-
tic conclusions drawn in this study were based on the Gibbs free-
energy profile of the entire catalytic cycle assessed at the B97-
D3(COSMO-RS)/(def2-TZVPPD+def2-TZVPD) level of approximation
[4] For biphenolate and binaphtholate rare-earth metal hydroamination
catalysts see: a) D. V. Gribkov, K. C. Hultzsch, F. Hampel, Chem. Eur. J.
2003, 9, 4796; b) D. V. Gribkov, F. Hampel, K. C. Hultzsch, Eur. J. Inorg.
Chem. 2004, 4091; c) D. V. Gribkov, K. C. Hultzsch, Chem. Commun.
2004, 730; d) D. V. Gribkov, K. C. Hultzsch, F. Hampel, J. Am. Chem. Soc.
2006, 128, 3748; e) A. L. Reznichenko, F. Hampel, K. C. Hultzsch, Chem.
Eur. J. 2009, 15, 12819; f) A. L. Reznichenko, H. N. Nguyen, K. C. Hultzsch,
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7855
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Angew. Chem. Int. Ed. 2010, 49, 8984; Angew. Chem. 2010, 122, 9168;
g) A. L. Reznichenko, K. C. Hultzsch, Organometallics 2013, 32, 1394;
h) X. Yu, T. J. Marks, Organometallics 2007, 26, 365.
man, J. Am. Chem. Soc. 1954, 76, 1899; d) R. Wegler, G. Pieper, Chem.
Ber. 1950, 83, 1.
[10] For selected examples of hydroamination of alkenes catalysed by alkali
metals see: a) M. Beller, C. Breindl, T. H. Riermeier, M. Eichberger, H.
Trauthwein, Angew. Chem. Int. Ed. 1998, 37, 3389; Angew. Chem. 1998,
110, 3571; b) D. Tzalis, C. Koradin, P. Knochel, Tetrahedron Lett. 1999, 40,
6193; c) A. Ates, C. Quinet, Eur. J. Org. Chem. 2003, 1623; d) V. Khedkar,
A. Tillack, C. Benisch, J.-P. Melder, M. Beller, J. Mol. Catal. A 2005, 241,
175; e) P. Horrillo Martínez, K. C. Hultzsch, F. Hampel, Chem. Commun.
2006, 2221; f) T. Ogata, A. Ujihara, S. Tsuchida, T. Shimizu, A. Kaneshige,
K. Tomioka, Tetrahedron Lett. 2007, 48, 6648; g) C. Quinet, P. Jourdain, C.
Hermans, A. Ates, I. Lucas, I. E. Markó, Tetrahedron 2008, 64, 1077; h) J.
Deschamp, J. Collin, J. Hannedouche, E. Schulz, Eur. J. Org. Chem. 2011,
3329.
[11] Reviews on the application of alkaline-earth metal catalysts: a) S.
Harder, Chem. Rev. 2010, 110, 3852; b) A. G. M. Barrett, M. R. Crimmin,
M. S. Hill, P. A. Procopiou, Proc. R. Soc. London Ser. A 2010, 466, 927;
c) M. R. Crimmin, M. S. Hill, Top. Organomet. Chem. 2013, 45, 191; d) S.
Kobayashi, Y. Yamashita, Acc. Chem. Res. 2011, 44, 58.
[12] For examples of alkaline-earth metal-based hydroamination catalysts
see: a) M. R. Crimmin, I. J. Casely, M. S. Hill, J. Am. Chem. Soc. 2005, 127,
2042; b) S. Datta, P. W. Roesky, S. Blechert, Organometallics 2007, 26,
4392; c) S. Datta, M. T. Gamer, P. W. Roesky, Organometallics 2008, 27,
1207; d) A. G. M. Barrett, M. R. Crimmin, M. S. Hill, P. B. Hitchcock, G.
Kociok-Kçhn, P. A. Procopiou, Inorg. Chem. 2008, 47, 7366; e) F. Buch, S.
Harder, Z. Naturforsch. B 2008, 63, 169; f) M. R. Crimmin, M. Arrowsmith,
A. G. M. Barrett, I. J. Casely, M. S. Hill, P. A. Procopiou, J. Am. Chem. Soc.
2009, 131, 9670; g) A. G. M. Barrett, C. Brinkmann, M. R. Crimmin, M. S.
Hill, P. Hunt, P. A. Procopiou, J. Am. Chem. Soc. 2009, 131, 12906; h) M.
Arrowsmith, M. S. Hill, G. Kociok-Kçhn, Organometallics 2009, 28, 1730;
i) M. Arrowsmith, A. Heath, M. S. Hill, P. B. Hitchcock, G. Kociok-Kçhn, Or-
ganometallics 2009, 28, 4550; j) P. Horrillo-Martínez, K. C. Hultzsch, Tetra-
hedron Lett. 2009, 50, 2054; k) M. Arrowsmith, M. S. Hill, G. Kociok-Kçhn,
Organometallics 2011, 30, 1291; l) J. Jenter, R. Kçppe, P. W. Roesky, Orga-
nometallics 2011, 30, 1404; m) M. Arrowsmith, M. R. Crimmin, A. G. M.
Barrett, M. S. Hill, G. Kociok-Kçhn, P. A. Procopiou, Organometallics 2011,
30, 1493; n) J. F. Dunne, D. B. Fulton, A. Ellern, A. D. Sadow, J. Am. Chem.
Soc. 2010, 132, 17680; o) A. Mukherjee, S. Nembenna, T. K. Sen, S. P.
Sarish, P. K. Ghorai, H. Ott, D. Stalke, S. K. Mandal, H. W. Roesky, Angew.
Chem. Int. Ed. 2011, 50, 3968; Angew. Chem. 2011, 123, 4054; p) S. R.
Neal, A. Ellern, A. D. Sadow, J. Organomet. Chem. 2011, 696, 228; q) J. S.
Wixey, B. D. Ward, Chem. Commun. 2011, 47, 5449; r) J. S. Wixey, B. D.
Ward, Dalton Trans. 2011, 40, 7693; s) T. D. Nixon, B. D. Ward, Chem.
Commun. 2012, 48, 11790; t) C. Brinkmann, A. G. M. Barrett, M. S. Hill,
P. A. Procopiou, J. Am. Chem. Soc. 2012, 134, 2193; u) B. Liu, T. Roisnel,
J.-F. Carpentier, Y. Sarazin, Angew. Chem. Int. Ed. 2012, 51, 4943; Angew.
Chem. 2012, 124, 5027; v) B. Liu, T. Roisnel, J.-F. Carpentier, Y. Sarazin,
Chem. Eur. J. 2013, 19, 2784; w) B. Liu, T. Roisnel, J.-F. Carpentier, Y. Sara-
zin, Chem. Eur. J. 2013, 19, 13445.
[5] Reviews on Group 4 metal-based catalysts: a) F. Pohlki, S. Doye, Chem.
Soc. Rev. 2003, 32, 104; b) I. Bytschkov, S. Doye, Eur. J. Org. Chem. 2003,
935; c) S. Doye, Synlett 2004, 1653; d) A. L. Odom, Dalton Trans. 2005,
225; e) R. Severin, S. Doye, Chem. Soc. Rev. 2007, 36, 1407; f) P. Eisen-
berger, L. L. Schafer, Pure Appl. Chem. 2010, 82, 1503. For selected ex-
amples of hydroamination of aminoalkenes catalysed by Group
4
metals see: g) P. D. Knight, I. Munslow, P. N. O’Shaughnessy, P. Scott,
Chem. Commun. 2004, 894; h) D. V. Gribkov, K. C. Hultzsch, Angew.
Chem. Int. Ed. 2004, 43, 5542; Angew. Chem. 2004, 116, 5659; i) J. A.
Bexrud, J. D. Beard, D. C. Leitch, L. L. Schafer, Org. Lett. 2005, 7, 1959;
j) H. Kim, P. H. Lee, T. Livinghouse, Chem. Commun. 2005, 5205; k) D. A.
Watson, M. Chiu, R. G. Bergman, Organometallics 2006, 25, 4731; l) C.
Müller, C. Loos, N. Schulenberg, S. Doye, Eur. J. Org. Chem. 2006, 2499;
m) M. C. Wood, D. C. Leitch, C. S. Yeung, J. A. Kozak, L. L. Schafer, Angew.
Chem. Int. Ed. 2007, 46, 354; Angew. Chem. 2007, 119, 358; n) A. L. Gott,
A. J. Clarke, G. J. Clarkson, P. Scott, Organometallics 2007, 26, 1729;
o) B. D. Stubbert, T. J. Marks, J. Am. Chem. Soc. 2007, 129, 6149; p) C.
Müller, R. Koch, S. Doye, Chem. Eur. J. 2008, 14, 10430; q) C. Müller, W.
Saak, S. Doye, Eur. J. Org. Chem. 2008, 2731; r) S. Majumder, A. L. Odom,
Organometallics 2008, 27, 1174; s) A. L. Reznichenko, K. C. Hultzsch, Or-
ganometallics 2010, 29, 24; t) L. E. N. Allan, G. J. Clarkson, D. J. Fox, A. L.
Gott, P. Scott, J. Am. Chem. Soc. 2010, 132, 15308; u) D. C. Leitch, R. H.
Platel, L. L. Schafer, J. Am. Chem. Soc. 2011, 133, 15453; v) K. Manna,
W. C. Everett, G. Schoendorff, A. Ellern, T. L. Windus, A. D. Sadow, J. Am.
Chem. Soc. 2013, 135, 7235; w) E. Chong, S. Qayyum, L. L. Schafer, R.
Kempe, Organometallics 2013, 32, 1858.
[6] Group 5 metal catalysts: a) A. L. Reznichenko, T. J. Emge, S. Audçrsch,
E. G. Klauber, K. C. Hultzsch, B. Schmidt, Organometallics 2011, 30, 921;
b) F. Zhang, H. Song, G. Zi, Dalton Trans. 2011, 40, 1547.
[7] Review on actinide-based catalysts: a) T. Andrea, M. S. Eisen, Chem. Soc.
Rev. 2008, 37, 550. For selected examples of actinide-catalysed hydroa-
mination of aminoalkenes, see: b) B. D. Stubbert, T. J. Marks, J. Am.
Chem. Soc. 2007, 129, 4253; c) E. M. Broderick, N. P. Gutzwiller, P. L. Diac-
onescu, Organometallics 2010, 29, 3242; d) C. E. Hayes, R. H. Platel, L. L.
Schafer, D. B. Leznoff, Organometallics 2012, 31, 6732. See also ref. [5o].
[8] Reviews on late transition metal-based catalysts: a) M. Beller, C. Breindl,
M. Eichberger, C. G. Hartung, J. Seayad, O. R. Thiel, A. Tillack, H. Trauth-
wein, Synlett 2002, 1579; b) J. F. Hartwig, Pure Appl. Chem. 2004, 76,
507; c) R. A. Widenhoefer, X. Han, Eur. J. Org. Chem. 2006, 4555; d) J. J.
Brunet, N. C. Chu, M. Rodriguez-Zubiri, Eur. J. Inorg. Chem. 2007, 4711;
e) K. D. Hesp, M. Stradiotto, ChemCatChem 2010, 2, 1192; f) J. Jenter, A.
Lühl, P. W. Roesky, S. Blechert, J. Organomet. Chem. 2010, 695, 406. Se-
lected recent examples: for the hydroamination of non-activated al-
kenes: g) C. F. Bender, R. A. Widenhoefer, J. Am. Chem. Soc. 2005, 127,
1070; h) A. Zulys, M. Dochnahl, D. Hollmann, K. Lçhnwitz, J.-S. Herr-
mann, P. W. Roesky, S. Blechert, Angew. Chem. Int. Ed. 2005, 44, 7794;
Angew. Chem. 2005, 117, 7972; i) X. Han, R. A. Widenhoefer, Angew.
Chem. Int. Ed. 2006, 45, 1747; Angew. Chem. 2006, 118, 1779; j) F. E. Mi-
chael, B. M. Cochran, J. Am. Chem. Soc. 2006, 128, 4246; k) C. Liu, J. F.
Hartwig, J. Am. Chem. Soc. 2008, 130, 1570; l) C. F. Bender, W. B.
Hudson, R. A. Widenhoefer, Organometallics 2008, 27, 2356; m) E. B.
Bauer, G. T. S. Andavan, T. K. Hollis, R. J. Rubio, J. Cho, G. R. Kuchenbeis-
er, T. R. Helgert, C. S. Letko, F. S. Tham, Org. Lett. 2008, 10, 1175; n) J.-W.
Pissarek, D. Schlesiger, P. W. Roesky, S. Blechert, Adv. Synth. Catal. 2009,
351, 2081; o) Z. Zhang, S. D. Lee, R. A. Widenhoefer, J. Am. Chem. Soc.
2009, 131, 5372; p) X. Shen, S. L. Buchwald, Angew. Chem. Int. Ed. 2010,
49, 564; Angew. Chem. 2010, 122, 574; q) K. D. Hesp, S. Tobisch, M. Stra-
diotto, J. Am. Chem. Soc. 2010, 132, 413; r) L. D. Julian, J. F. Hartwig, J.
Am. Chem. Soc. 2010, 132, 13813; s) Z. Liu, H. Yamamichi, S. T. Madrahi-
mov, J. F. Hartwig, J. Am. Chem. Soc. 2011, 133, 2772; t) C. S. Sevov, J.
Zhou, J. F. Hartwig, J. Am. Chem. Soc. 2012, 134, 11960; u) M. Rodriguez-
Zubiri, C. Baudequin, A. BØthegnies, J.-J. Brunet, ChemPlusChem 2012,
77, 445; v) E. Bernoud, P. OuliØ, R. Guillot, M. Mellah, J. Hannedouche,
Angew. Chem. Int. Ed. 2014, 53, 4930; Angew. Chem. 2014, 126, 5030.
[9] a) J. Seayad, A. Tillack, C. G. Hartung, M. Beller, Adv. Synth. Catal. 2002,
344, 795; b) R. D. Closson, J. P. Napolitano, G. G. Ecke, A. Kolka, J. Org.
Chem. 1957, 22, 646; c) B. W. Howk, E. L. Little, S. L. Scott, G. M. Whit-
[13] a) X. Zhang, T. J. Emge, K. C. Hultzsch, Organometallics 2010, 29, 5871;
b) X. Zhang, T. J. Emge, K. C. Hultzsch, Angew. Chem. Int. Ed. 2012, 51,
394; Angew. Chem. 2012, 124, 406.
[14] D. Seyferth, Organometallics 2009, 28, 1598.
[15] a) P. W. Roesky, T. E. Müller, Angew. Chem. Int. Ed. 2003, 42, 2708; Angew.
Chem. 2003, 115, 2812; b) K. C. Hultzsch, Adv. Synth. Catal. 2005, 347,
367; c) K. C. Hultzsch, Org. Biomol. Chem. 2005, 3, 1819; d) I. Aillaud, J.
Collin, J. Hannedouche, E. Schulz, Dalton Trans. 2007, 5105; e) G. Zi,
Dalton Trans. 2009, 9101; f) A. L. Reznichenko, K. C. Hultzsch, in Chiral
Amine Synthesis: Methods, Developments and Applications (Ed: T.
Nugent), Wiley-VCH, Weinheim, 2010, p.341; g) G. Zi, J. Organomet.
Chem. 2011, 696, 68; h) J. Hannedouche, J. Collin, A. Trifonov, E. Schulz,
J. Organomet. Chem. 2011, 696, 255; i) A. L. Reznichenko, A. J. Nawara-
Hultzsch, K. C. Hultzsch, Top. Curr. Chem. 2014, 343, 191.
[16] For studies on the application of phenoxyamine alkaline-earth com-
plexes as initiators in the polymerization of cyclic esters and lactide,
see: a) J. Ejfler, M. Kobyłka, L. B. Jerzykiewicz, P. Sobota, Dalton Trans.
2005, 2047; b) Y. Sarazin, R. H. Howard, D. L. Hughes, S. M. Humphrey,
M. Bochmann, Dalton Trans. 2006, 340; c) D. J. Darensbourg, W. Choi,
C. P. Richers, Macromolecules 2007, 40, 3521; d) Z. Zheng, G. Zhao, R.
Fablet, M. Bouyahyi, C. M. Thomas, T. Roisnel, O. Casagrande, Jr., J.-F.
Carpentier, New J. Chem. 2008, 32, 2279; e) M. G. Davidson, C. T. O’Hara,
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7856
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
M. D. Jones, C. G. Keir, M. F. Mahon, G. Kociok-Kçhn, Inorg. Chem. 2007,
46, 7686; f) V. Poirier, T. Roisnel, J.-F. Carpentier, Y. Sarazin, Dalton Trans.
2009, 9820; g) W.-C. Hung, C.-C. Lin, Inorg. Chem. 2009, 48, 728; h) J.
Ejfler, K. Krauzy-Dziedzic, S. Szafert, L. B. Jerzykiewicz, P. Sobota, Eur. J.
Inorg. Chem. 2010, 3602; i) Y. Sarazin, V. Poirier, T. Roisnel, J.-F. Carpenti-
er, Eur. J. Inorg. Chem. 2010, 3423; j) B. Liu, T. Roisnel, J.-P. GuØgan, J.-F.
Carpentier, Y. Sarazin, V. Poirier, T. Roisnel, J.-F. Carpentier, Chem. Eur. J.
2012, 18, 6289.
[30] D. S. C. Black, J. E. Doyle, Austr. J. Chem. 1978, 31, 2247.
[31] T. Kondo, T. Okada, T. Mitsudo, J. Am. Chem. Soc. 2002, 124, 186.
[32] The primary KIE has been estimated from the deviation of zero-point vi-
brational energies and partition function values for S and [D2]-S sub-
strates. Zero curvature tunnelling has been accounted for by the
Skodje and Truhlar approximation, see: a) R. T. Skodje, D. G. Truhlar,
B. C. Garrett, J. Phys. Chem. 1981, 85, 3019; the accuracy of computa-
tionally predicted kH/kD values for hydrogen transfer processes is limited
by transition-state theory approximation, semi-classical tunnelling cor-
rections and rigid rotator, harmonic oscillator approximation for vibra-
tional frequencies, see: b) D. H. Lu, D. Maurice, D. G. Truhlar, J. Am.
Chem. Soc. 1990, 112, 6206.
[17] The synthesis and X-ray structural analysis of complex 1a is reported
in: A. R. F. Cox, V. C. Gibson, E. L. Marshall, A. J. P. White, D. Yeldon,
Dalton Trans. 2006, 5014.
[18] For DFT studies on the rare-earth metal-catalysed hydroamination, see:
a) A. Motta, G. Lanza, I. L. Fragalà, T. J. Marks, Organometallics 2004, 23,
4097; b) A. Motta, I. L. Fragalà, T. J. Marks, Organometallics 2006, 25,
5533; c) S. Tobisch, J. Am. Chem. Soc. 2005, 127, 11979; d) S. Tobisch,
Chem. Eur. J. 2005, 11, 6372; e) S. Tobisch, Chem. Eur. J. 2006, 12, 2520;
f) S. Tobisch, Chem. Eur. J. 2007, 13, 9127; g) S. Tobisch, Chem. Eur. J.
2010, 16, 13814; h) S. Tobisch, Dalton Trans. 2012, 41, 9182; i) S. Tobisch,
Chem. Eur. J. 2014, 20, 8988.
[33] R. G. Parr, W. Yang, Density-Functional Theory of Atoms and Molecules,
Oxford University Press, Oxford, 1989.
[34] a) R. Ahlrichs, M. Bär, M. Häser, H. Horn, C. Kçlmel, Chem. Phys. Lett.
1989, 162, 165; b) O. Treutler, R. Ahlrichs, J. Chem. Phys. 1995, 102, 346;
c) R. Ahlrichs, F. Furche, C. Hättig, W. Klopper, M. Sierka, F. Weigend,
TURBOMOLE, version 6.0, University of Karlsruhe, 2009, http://www.tur-
bomole.com.
[19] S. Tobisch, Inorg. Chem. 2012, 51, 3786.
[20] S. Tobisch, Chem. Eur. J. 2011, 17, 14974.
[35] a) S. Grimme, J. Comput. Chem. 2006, 27, 1787; b) A. D. Becke, J. Chem.
Phys. 1997, 107, 8554.
[21] Substrate association and dissociation at the various intermediates is
presumably facile and rapid equilibria are expected; examination by
a linear-transit approach gave no indication that such processes are as-
sociated with a significant enthalpic barrier.
[36] a) O. Vahtras, J. Almlçf, M. W. Feyereisen, Chem. Phys. Lett. 1993, 213,
514; b) K. Eichkorn, O. Treutler, H. Öhm, M. Häser, R. Ahlrichs, Chem.
Phys. Lett. 1995, 242, 652.
[37] a) S. Grimme, J. Anthony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010, 132,
154104; b) S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 2011, 32,
1456; c) http://toc.uni-muenster.de/DFTD3/getd3.html.
[22] See the Supporting Information for more detail.
[23] The prevalent [{ONN}Mg(NHR)·(NH2R)] form C3a1’·S, featuring a chelating
amidoalkene unit, of the competent catalyst complex (with the appro-
priate number of substrate S or cycloamine P molecules) has been
chosen as reference for relative free energies (given in kcalmolÀ1).
[24] The tBu, SiPh3 and methyl groups of the phenoxyamine {ONN}Mg
ligand backbone are in grey together with a truncated display of the
aminoalkene in order to enhance the visualisation of crucial structural
aspects.
[25] Please note that the thermodynamic force of 7.8 kcalmolÀ1 for S!P
cyclisation has been subtracted from C5, TS[C5–C6·S] and C6·S in
Figure 16; The most stable conformer of C5 and C6·S is included in
Figure 16.
[38] a) F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 2005, 7, 3297; b) F.
Weigend, Phys. Chem. Chem. Phys. 2006, 8, 1057.
[39] a) A. Schäfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1994, 100, 5829; b) K.
Eichkorn, F. Weigend, O. Treutler, R. Ahlrichs, Theor. Chem. Acc. 1997, 97,
119; c) A. Schäfer, C. Huber, R. Ahlrichs, J. Chem. Phys. 1992, 97, 2571.
[40] a) A. Klamt, G. Schüürmann, J. Chem. Soc. Perkin Trans. 2 1993, 799;
b) A. Klamt, J. Phys. Chem. 1995, 99, 2224; c) F. Eckert, A. Klamt, AIChE J.
2002, 48, 369.
[41] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098; b) J. P. Perdew, Phys. Rev. B
1986, 33, 8822.
[42] Y. Okuno, Chem. Eur. J. 1997, 3, 212.
[26] R. R. Schrock, J. Organomet. Chem. 1976, 122, 209.
[43] For further details, see http://www.struked.de.
[27] A. N. Thadani, Y. Huang, V. H. Rawal, Org. Lett. 2007, 9, 3873.
[28] C. K. Williams, L. E. Breyfogle, S. K. Choi, W. Nam, J. Victor, G. Young,
M. A. Hillmyer, W. B. Tolman, J. Am. Chem. Soc. 2003, 125, 11350.
[29] Y. Tamaru, M. Hojo, H. Higashimura, Z. Yoshida, J. Am. Chem. Soc. 1988,
110, 3994.
Received: December 12, 2014
Revised: March 9, 2015
Published online on April 13, 2015
Chem. Eur. J. 2015, 21, 7841 – 7857
www.chemeurj.org
7857
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim