A Highly Effective Ruthenium System for the Catalyzed Dehydrogenative Cyclization of Amine-Boranes to Cyclic Boranes under Mild Conditions

Article abstract of DOI:10.1002/chem.201501569

We recently disclosed a new ruthenium-catalyzed dehydrogenative cyclization process (CDC) of diamine-monoboranes leading to cyclic diaminoboranes. In the present study, the CDC reaction has been successfully extended to a larger number of diamine-monoboranes (4-7) and to one amine-borane alcohol precursor (8). The corresponding NB(H)N- and NB(H)O-containing cyclic diaminoboranes (12-15) and oxazaborolidine (16) were obtained in good to high yields. Multiple substitution patterns on the starting amine-borane substrates were evaluated and the reaction was also performed with chiral substrates. Efforts have been spent to understand the mechanism of the ruthenium CDC process. In addition to a computational approach, a strategy enabling the kinetic discrimination on successive events of the catalytic process leading to the formation of the NB(H)N linkage was performed on the six-carbon chain diamine-monoborane 21 and completed with a ¹⁵NNMR study. The long-life bis-σ-borane ruthenium intermediate 23 possessing a reactive NHMe ending was characterized in situ and proved to catalyze the dehydrogenative cyclization of 1, ascertaining that bis σ-borane ruthenium complexes are key intermediates in the CDC process. Mild-mannered cyclization: The ruthenium-catalyzed dehydrogenative cyclization (CDC) process has been investigated by using a variety of amine-borane substrates (see figure). The reaction leads to the formation of the corresponding NB(H)N- and NB(H)O-containing cyclic aminoboranes and involves the formation of a bis-σ-borane ruthenium intermediate complex.

Full text of DOI:10.1002/chem.201501569

DOI: 10.1002/chem.201501569
Full Paper
&
Boranes
A Highly Effective Ruthenium System for the Catalyzed
Dehydrogenative Cyclization of Amine–Boranes to Cyclic Boranes
under Mild Conditions
Christopher J. Wallis,[a, b] Gilles Alcaraz,*[a, b] Alban S. Petit,[c] Amalia I. Poblador-Bahamonde,[c]
Eric Clot,[c] Christian Bijani,[a, b] Laure Vendier,[a, b] and Sylviane Sabo-Etienne[a, b]
Abstract: We recently disclosed a new ruthenium-catalyzed
dehydrogenative cyclization process (CDC) of diamine–
monoboranes leading to cyclic diaminoboranes. In the pres-
ent study, the CDC reaction has been successfully extended
to a larger number of diamine–monoboranes (4–7) and to
one amine–borane alcohol precursor (8). The corresponding
NB(H)N- and NB(H)O-containing cyclic diaminoboranes (12–
15) and oxazaborolidine (16) were obtained in good to high
yields. Multiple substitution patterns on the starting amine–
borane substrates were evaluated and the reaction was also
performed with chiral substrates. Efforts have been spent to
understand the mechanism of the ruthenium CDC process.
In addition to a computational approach, a strategy enabling
the kinetic discrimination on successive events of the cata-
lytic process leading to the formation of the NB(H)N linkage
was performed on the six-carbon chain diamine–monobor-
ane 21 and completed with a 15N NMR study. The long-life
bis-s-borane ruthenium intermediate 23 possessing a reac-
tive NHMe ending was characterized in situ and proved to
catalyze the dehydrogenative cyclization of 1, ascertaining
that bis s-borane ruthenium complexes are key intermedi-
ates in the CDC process.
Introduction
In 2007, we disclosed the unprecedented symmetrical coor-
dination of a monosubstituted borane (RBH2) to a ruthenium
center[4] involving the geminal s-BÀH bonds of the borane in
two three-center two-electron bond interactions with the
metal.[5] Three years later, we reported that the bis(dihydrogen)
complex [RuH2(h2-H2)2(PCy3)2] (I) (Figure 1) led to the stoichio-
metric dehydrogenation of amine–boranes H3BÀNMe3ÀnHn (n=
1–3), at room temperature, with formation of the correspond-
ing bis-s-BÀH aminoborane ruthenium complexes [RuH2(h2:h2-
The homogeneous transition-metal-catalyzed dehydrogenation
of amine–boranes has produced a wide range of academic re-
search in an attempt to identify suitable catalytic candidates,
as well as to determine their mechanism of action.[1] The dehy-
drogenation of ammonia–borane, for example, is still regarded
as one of the most interesting models for chemical hydrogen
storage.[2] Many effective homogeneous catalytic systems have
been discovered, and it is now well established that both the
type of the transition-metal catalyst and the nature of the
amine–borane unit chosen determine the identity of the prod-
uct resulting from the dehydrogenation process.[1b,2] Most sys-
tems to date produce the corresponding polymeric aminobor-
anes, cyclic borazines, and/or polyborazylenes, precursors of
BN-containing materials.[3]
[a] Dr. C. J. Wallis, Dr. G. Alcaraz, Dr. C. Bijani, Dr. L. Vendier,
Dr. S. Sabo-Etienne
CNRS, LCC (Laboratoire de Chimie de Coordination),
205 route de Narbonne, BP 44099, 31077 Toulouse Cedex 4 (France)
E-mail: gilles.alcaraz@lcc-toulouse.fr
[b] Dr. C. J. Wallis, Dr. G. Alcaraz, Dr. C. Bijani, Dr. L. Vendier,
Dr. S. Sabo-Etienne
UniversitØ de Toulouse, UPS, INPT, 31077 Toulouse Cedex 4 (France)
[c] Dr. A. S. Petit, Dr. A. I. Poblador-Bahamonde, Dr. E. Clot
Institut Charles Gerhardt, UniversitØ de Montpellier,
CNRS 5253, cc 1501, Place Eug›ne Bataillon,
34095 Montpellier Cedex 5 (France)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201501569.
Figure 1. Stoichiometric and catalytic ruthenium dehydrogenation of
amine–boranes.
Chem. Eur. J. 2015, 21, 13080 – 13090
13080
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
H2BNHnÀ1Me3Àn)(PCy3)2] (A) (Figure 1).[6] We clearly showed
1) the ability of ruthenium to retain the BÀN unit along the el-
ementary steps of the amine–borane dehydrogenation path-
ways, and 2) the bis-s-BÀH coordination mode to be a useful
tool for the stabilization of monomeric aminoboranes in the
coordination sphere of the metal,[7] enabling even the trapping
of the simplest and elusive prototypical H2BÀNH2 elementary
unit (Figure 1).[1a,6]
Table 1. Ruthenium CDC of amine–borane substrates.
Entry
Substrate
Product
t
Yield
[%][c]
[h][b]
1[a]
2[a]
3[a]
4
3
8
25
55
25
88
72
As a part of our research program focusing on BÀH bond ac-
tivation, we were interested to further explore the reactivity of
the bis-s-BÀH coordination mode, based on an approach remi-
niscent of the intramolecular heterolytic cleavage process of H2
taking place in the first coordination sphere of a metal in the
presence of a nucleophile.[8] We were particularly interested in
investigating the potential intramolecular interaction of the ac-
tivated s-bound BÀH units towards a Lewis base (LB) moiety
located away from the nitrogen atom of the aminoborane unit
(B). We found that diamine–monoboranes possessing a secon-
dary amine end group were converted into the corresponding
cyclic diaminoborane C through a catalyzed dehydrogenative
cyclization (CDC) process with I as a catalyst precursor
(Figure 1).[9] In the case of a pendant tertiary amine group, no
cyclization occurred; the reaction was stoichiometric, produc-
ing solely the bis-s-BÀH aminoborane ruthenium complex D[9]
(Figure 1). Because the outcome of the reaction is influenced
by the substitution pattern of the remote amine group,
a deeper investigation of the reactivity of a series of diversely
functionalized amine–boranes was undertaken. The aim is to
better delineate the synthetic potential of this reaction and to
provide mechanistic insights, in the continuation of our pre-
liminary results.
3
16
72
5
6
30
78
7
30
24
78
42
8[d]
[a] See previous communication.[9] [b] Time for 100% conversion of the
substrate determined by 11B NMR spectroscopy. [c] Yield of the isolated
product. [d] Monomer–dimer equilibrium observed for 16.[11]
Results and Discussion
We have already disclosed the ability of [RuH2(h2-H2)2(PCy3)2] (I)
to be a highly effective and robust catalyst for the CDC trans-
formation of three diamine–monoboranes into their cyclic di-
aminoborane derivatives (Table 1, entries 1–3). Our efforts here
focus on expanding the substrate scope and showing the ver-
satile nature of the catalyst towards a variety of substrates. The
diamine–monoborane substrates 1–7 (Table 1) used for the
catalytic studies were synthesized by the stoichiometric reac-
tion of BH3·SMe2 with the corresponding diamine at À788C in
THF. The products were mostly isolated as colorless oils and
fully characterized by NMR spectroscopy. Compounds 6 and 7,
synthesized from the starting enantiopure diamines, were ob-
tained in a 9:1 diastereoisomeric integration ratio due to the
presence of a chiral nitrogen atom generated by quaterniza-
tion with BH3.
Scheme 1. Synthesis of substrate 8.
Substrates 4 and 5 were chosen to further evaluate the influ-
ence of the steric bulk of the substituent on the nitrogen
atoms. The change from a methyl to an isopropyl group had
already been shown to more than double the reaction time,
from 3 to 8 h (Table 1, entries 1 and 2). Thus, increasing further
the steric bulk to a tert-butyl group accordingly increases the
reaction time to 16 h (Table 1, entry 4). Surprisingly, however,
the benzyl-substituted substrate (Table 1, entry 5), which is
sterically less encumbered than the tert-butyl analogue, does
not follow the same correlation, with a 72 h long CDC reaction
time for 100% conversion. We speculate that the increased re-
action time could be due to the weaker basicity of the N-
benzyl-substituted amine end group.[12] In the case of di-
amine–monoboranes 6 and 7, the CDC transformation took
30 h to reach complete consumption of substrate, highlighting
the influence of the backbone rigidity in the cyclization pro-
cess for these substrates. The corresponding diazaborolidines
Synthesis of 8 was performed from the enantiopure (S)-(À)-
a,a-diphenyl-2-pyrrolidinemethanol by adding two equivalents
of BH3·THF followed by a hydrolytic work-up to restore the al-
cohol moiety while still guarding the amine–borane group
(Scheme 1).[10]
By using the same reported conditions,[9] the CDC reactions
were conducted with substrate 1–8 in THF, in the presence of
a 2.5 mol% of I as a catalyst (Table 1).
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13081
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
14 and 15 were isolated in 78% yield and one single signal
was observed, respectively, in the 11B NMR spectrum. The spe-
cific optical rotation for each diazaborolidine was determined
(14:½aŠ20 =À354Æ9 (c=2.76, CH2Cl2), 15: ½aŠ20 = +352Æ9 (c=
moiety, in the solid state. However, the H NMR study of D at
room temperature in a [D8]THF solution revealed the magnetic
nonequivalence of the terminal hydrides (d=À11.94,
À12.27 ppm).[9] This feature was not observed in the
case of the previously reported complex [RuH2((h2:h2-
H2BNHMe)(PCy3)2][6] with two different substituents attached to
the terminal nitrogen atom. This difference is presumably be-
cause of a weak barrier to rotation around the BÀN bond, ren-
dering the hydrogen atoms surrounding the ruthenium center
equivalent. The assumption of an intramolecular interaction in-
creasing the rotational barrier, likely to eliminate magnetic
equivalence of the hydride ligands (as depicted in Figure 2),
needed to be carefully examined. We checked that point by
changing the nature of the peripheral ending group and by
modifying the length of the diamine backbone.
D
D
2.83, CH2Cl2)) clearly showing they are enantiomers and that
the chirality is preserved during the process.
To expand the scope of the CDC, we examined the nature of
the pendant group by changing the amino group to an alco-
hol function. Under the same conditions and starting from 8,
the chiral oxazaborolidine 16 was produced cleanly, and
showed the reported dynamic behavior in solution with inter-
conversion between its monomeric and dimeric forms at ambi-
ent temperature in THF.[11] With this study, we have illustrated
the versatility of complex [RuH2(h2-H2)2(PCy3)2] (I) for a range of
substrates with different steric and electronic properties in
CDC transformations. The reactions remain clean when moni-
tored by NMR spectroscopy and the cyclized products were
quantitatively obtained and isolated in very good to moderate
yields when volatile.
The stoichiometric reaction of I with amine–borane adducts
17 and 18 led to the corresponding bis-s-borane ruthenium
complexes 19 and 20, respectively, in high yields (Scheme 2).
1
The H NMR data were collected and a similar trend was ob-
As reported in our previous communication, complex I
serves as a precatalyst and is also identified as the catalyst rest-
ing state of the reaction,[9] I being fully regenerated during the
CDC process presumably as a result of evolution of dihydro-
gen. Despite the absence of observable intermediates,[13] we
speculated the reaction mechanism would involve the initial
formation of a corresponding bis-s-borane ruthenium complex
with concomitant or subsequent creation of an interaction be-
tween the peripheral ending group and the bis-s-coordinated
borane moiety. In this respect, several scenarios can be envis-
aged, as depicted in Figure 2, including a Lewis acid–base in-
teraction (E), the nucleophilic cleavage of the s-coordinated
BÀH bond (F) or dihydrogen bonding (G)[14] as the prelude to
the final dehydrogenation step.
served for the terminal hydrides in solution at room tempera-
ture (19: d=À12.12, À12.40; 20: d=À11.92, À12.38 ppm).
Deeper 1H NMR investigations of complexes D, 19, and 20
showed each set of signals in the hydride zone, centered at
about d=À7 and À12 ppm, respectively, are averaged upon
heating with a decoalescence temperature close to 310 K (see
the Supporting Information, Figures S2, S6, and S11). Typically,
in the case of 19, the 2D 1H-EXSY (the Supporting Information,
Figure S8) and 2D 1H-ROESY (the Supporting Information, Fig-
ure S9) experiments carried out at 243 K showed that the ter-
minal hydrides are exchanging together. The same feature was
observed for the boron-bonded hydrides. Consequently, the
magnetic nonequivalence observed at room temperature for
the four hydrides in these complexes is more probably due to
a slow rotation about the BÀN bond on the NMR timescale,[15]
than the result of an intramolecular interaction as depicted in
Figure 2. Different scenarios for the final dehydrogenation step in the ruthe-
nium CDC transformation.
Our efforts then focused on trying to isolate the postulated
bis-s-BÀH aminoborane-ruthenium complex intermediate in
the CDC reaction to track a possible intramolecular interaction
as described above. In the preliminary communication, we
showed the stoichiometric reaction of N,N’,N’-trimethyl-
ethylenediamine–monoborane with I led to the formation
of
the
stable
bis
s-BÀH
ruthenium
complex
[RuH2(h2:h2:H2BN(Me)CH2CH2NMe2)(PCy3)2] (D) (Figure 1). The X-
ray diffraction structure showed that the diamine backbone in
D adopts an open chain structure without any interaction be-
tween the pendant NMe2 group and the ligated borane
Scheme 2. Synthesis of the bis-s-borane ruthenium complexes 19 and 20.
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13082
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 2. Experimentally, the Eyring and Arrhenius plots of the
rate constants (the Supporting Information, Figures S3, S7, and
1
S12) obtained by H{31P:11B} NMR line shape simulations from
the two sets of hydride signals over a significant range of tem-
perature, allowed us to determine a set of thermodynamic pa-
rameters (DG°coal:, DH°, and DS°) and to estimate the activation
energy (Ea), respectively, for each complex (Table 2). This unpre-
cedented study on the BÀN rotational barriers in bis-s-amino-
borane ruthenium complexes is discussed further below.
Table 2. Thermodynamic parameters and activation energy for com-
plexes D, 19, 20, and 23.
Ea
DH°
[kJmolÀ1
DS°
[JmolÀ1 KÀ1
]
G°coal: [kJmolÀ1
(at T [K])
]
[kJmolÀ1
]
]
D
61.0Æ5.4 (BH2)
61.2Æ5.2 (RuH2)
55.3Æ3.8 (BH2)
54.9Æ2.0 (RuH2)
53.8Æ1.7 (BH2)
54.1Æ2.4 (RuH2)
51.8Æ1.7 (BH2)
52.5Æ1.2 (RuH2)
58.7Æ5.4
58.8Æ5.3
52.9Æ3.7
52.5Æ2.0
51.4Æ1.8
51.7Æ2.3
49.4Æ1.7
50.0Æ1.2
À2.5Æ19
À2.0Æ19
À21Æ13
À22Æ5
À29Æ6
À29Æ8
À38Æ6
À36Æ4
59.4 (298)
59.8 (313)
59.1 (298)
59.1 (308)
60.3 (308)
60.8 (318)
60.8 (308)
61.4 (318)
19
20
23
X-ray crystallography characterization of complexes 19 and
20 (at 110 K) revealed that the peripheral ending groups are
again positioned far away from the coordinated borane
moiety, in the solid state (Figure 3).
In addition to the amine–borane moiety, the structural re-
quirements of the substrate necessary for the achievement of
the CDC transformation include the presence of a remote
protic Lewis base. In the absence of any detected intramolecu-
lar interaction between the h2:h2-coordinated BH2 fragment
and the methoxy group in 19 or the dimethylamino ending in
D and 20, the involvement of a bis-s borane ruthenium com-
plex as a transient intermediate from which one of the scenar-
ios depicted in Figure 2 could be developed to produce
Figure 3. X-ray structures of complexes 19 and 20. The hydrogen atoms not
associated with the metal centers have been omitted for clarity. Ellipsoids
set at 30% probability.
a
cyclic diaminoborane, remained questionable. To gain
a better insight in this mechanism, we envisioned a strategy
that would allow the accumulation and thus the detection of
a key catalytic intermediate. We thus turned our attention to-
wards a substrate allowing a kinetic discrimination of the suc-
cessive events leading to the formation of the NB(H)N linkage.
This strategy was conducted by lengthening the carbon chain
of the diamine–monoborane to six carbon atoms. We found
that 21, in the presence of 2.5 mol% of I, was fully consumed
after 48 h at room temperature in [D8]THF. The NB(H)N linkage
resulting from the dehydrogenative process was clearly identi-
fied with the presence of a characteristic signal at about d=
29 ppm (21: d=À15 ppm; BH3) for the boron (the Supporting
Information, Figure S13) and at d=3.75 ppm (21: d=
1.49 ppm; BH3) for the hydrogen atom bonded to boron (the
Supporting Information, Figure S14), in the 11B{1H} and
1H{11B} NMR spectra, respectively. The presence of a signal at
about d=4 ppm in the 11B{1H} NMR spectrum also indicates
the presence of cyclodiborazane 22 (Scheme 3) in equilibrium
with its monomeric aminoborane form (d=38 ppm) upon
Scheme 3. Reaction of diamine-monoboranes 21 and 1 with I and 23, re-
spectively.
heating (the Supporting Information, Figure S15). In the ab-
sence of I, no dehydrogenation takes place and 21 remained
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13083
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
unchanged after stirring for 5 days in the same conditions.
Monitoring the reaction by NMR spectroscopy in the presence
of I under the reported CDC conditions, we observed the com-
plete consumption of catalyst I during the process and the
generation of a species resembling a bis-s-BÀH diaminoborane
ruthenium complex, which remained at the end of the catalytic
process (the Supporting Information, Figures S16 and S17).
This result differs quite significantly compared with the other
diamine–monoborane substrates for which spectroscopic data
suggest that the ruthenium catalyst I always remains intact
and reusable at the end of the CDC process. To clearly charac-
terize this species, we treated I with a sub- or super-stoichio-
metric amount of 21 at room temperature. Under these condi-
tions, complex 23 could be generated and appeared stable
enough to be spectroscopically characterized in solution
(Scheme 3). Under super-stoichiometric conditions (I/21=
1:1.5) in [D8]THF at RT, compound 23 was produced as the
major product along with minor amount of the CDC product
and traces of cyclodiborazane 22 according to the 11B NMR
spectrum (the Supporting Information, Figure S18). The broad
signal at d=47.5 ppm is consistent with the chemical shifts of
previously reported bis-s-aminoborane ruthenium complexes.
The 31P{1H} NMR spectrum exhibits two overlapping singlets at
d=79.22 and 79.20 ppm that split upon heating, presumably
as the result of two conformers (the Supporting Information,
on boron,[4] together with the donation from the lone pair of
the adjacent nitrogen atom.[7,17] The weakly negative activation
entropy value is indicative of a more ordered transition state in
the process. These relatively small values are indicative of a dy-
namic intramolecular process resulting from the rotation
around the BÀN bond rather than a decoordination–coordina-
tion pathway of the aminoborane.
A comparative 15N NMR study of complex 23 with diamine
24, amine–boranes 21 and 25, and complex 26, which possess-
es a carbon chain of similar length attached to the nitrogen
(the Supporting Information, Figures S23–S29), enabled the un-
ambiguous characterization of the pendant NHMe moiety at
d=À358 ppm (Table 3), hence validating the structure of 23 in
solution.
Table 3. Comparative study of compounds 21 and 23–26 by 15N NMR
spectroscopy.
Compounds
d15N [ppm]
À351(1J15NH =67 Hz) (H3BNHMe)
À357 (NHMe)
21
23
À299 (H2BN)
À358 (NHMe)
1
Figure S19). In the H NMR spectrum at 298 K, two sets of sig-
24
25
À358 (NHMe)
nals at d=À7.3 and À12.7 ppm are observed in the hydride
zone in a 1:1 integration ratio (the Supporting Information, Fig-
ure S20). The more shielded terminal hydrides and the broad
BH signals in 23 are averaged upon heating, allowing the de-
termination of the thermodynamic parameters that are pre-
sented in Table 2 (the Supporting Information, Figures S21 and
S22). The values obtained for complex 23 are similar to those
of D, 19, and 20. This result is consistent with a slow rotation
around the BÀN bond on the NMR timescale, exchanging the
terminal hydrides, on the one hand, and the BH hydrides on
the other.
À350 (1J15NH =71 Hz) (H3BNHMe)
26
À299 (H2BN)
On that basis, we then showed that the transformation of
1 into 9 was catalyzed by 2.5 mol% of 23 at room temperature
(Scheme 3). After 24 h, in addition to the formed diazaboroli-
dine 9, the presence of 23 was, here again, clearly identified
The activation energy values for these complexes are of sim-
ilar magnitude (Table 2) but smaller than those for free H2BÀ by 1H and 31P{1H} NMR spectroscopy, along with traces of I
NH2, computed to be 141.4 kJmolÀ1 for the rotation barrier
with an adjacent trigonal planar nitrogen atom, and
125.1 kJmolÀ1 after relaxation of the nitrogen to a tetrahedral
geometry, as the estimation of the p contribution to the BÀN
bond.[16] In our study, the rotational barrier is about 50–
generated during the CDC process (the Supporting Informa-
tion, Figures 30–33).
In light of these results, it seems now realistic to consider
the bis-s-borane ruthenium complexes as likely intermediates
in the CDC reaction pathway. The formation of the cyclodibor-
62 kJmolÀ1, and about 80 kJmolÀ1 smaller than that in H2BÀ azane 22 observed at the end of the CDC process from 21 is
NH2, illustrating a reduced BÀN p-bonding character in bis-s-
(BÀH) aminoborane complexes.
consistent with this hypothesis. The presence of 22 in solution
would result from the displacement of the coordinated amino-
borane species by the in situ-generated dihydrogen as a com-
peting ligand at ruthenium, and from its subsequent dimeriza-
tion in solution (Scheme 3).
The structure of the transition state for BÀN rotation in bis-
s-BÀH aminoborane ruthenium complexes should result from
a 908 rotation around the BÀN bond together with a decrease
in the BÀN p-bonding, increase in the BÀN bond length, and
pyramidalization of the nitrogen atom with the lone pair or-
thogonal to the boron pz orbital. However, no particular de-
shielding of boron was observed by 11B NMR spectroscopy for
DFT (B3PW91) calculations were performed next to obtain
insights into the CDC mechanism and the rotational process in
the bis-s adducts (see the Computational Details). The
bis-s complexes D and 19 have been computed and the calcu-
lated geometries (D-calc and 19-calc) are in excellent agree-
ment with the experimental data (see Table S1 in the Support-
ing Information). The transition-state structures (TSrot-D-calc
complexes
D (d=47.4 ppm), 19 (d=47.6 ppm), 20 (d=
47.4 ppm), and 23 (d=47.5 ppm). This probably results from
the synergistic p back-donation from Ru to the empty p orbital
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13084
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
and TSrot-19-calc) associated with the rotation around the
BÀN bond have been located on the potential energy surface
(PES).[18] As expected, the transition state corresponds to an
almost 908 value for the dihedral angle between the BH2
plane, which stays coordinated in a bis-s fashion to Ru, and
the NMeR plane (R=CH2CH2NMe2, TSrot-D-calc; R=
CH2CH2OMe, TSrot-19-calc).
Upon rotation, the RuÀB bond length decreases by about
0.07 Š and the BÀN bond lengthens by about 0.02 Š (Table 4).
This is the result of competition of p-donation into the empty
p AO on boron from both the occupied d AO on ruthenium
and the nitrogen lone pair. In the TS for rotation, the loss of p-
donation from nitrogen (longer BÀN bond) is compensated by
increased p-donation from ruthenium (shorter RuÀB). Interest-
ingly, the hybridization at nitrogen is only slightly altered in
the transition state and remains close to trigonal planar
(Table 4).
temperature and, therefore, with the observation of equivalent
hydrides.
The NMR experiments and the calculations on the bis-s ad-
ducts did not evidence any significant interaction between the
remote Lewis base (NMe2, NHMe, or OMe) and the boron
atom. The experiment with 21 indicated that a bis-s adduct is
an intermediate on the pathway for dehydrogenative cycliza-
tion (DC) of amine–borane. However, the experiment with
1 showed that the lifetime of this bis-s adduct is not long
enough to be observed experimentally when the tail is short
(here CH2CH2NMeH). The bis-s adduct is thus expected to be
formed more slowly than its subsequent transformation to the
cyclized product, yet the bis-s adduct is thermodynamically fa-
vored over the bis-dihydrogen starting catalyst. This is con-
firmed computationally with the bis-s adduct DH-calc, com-
puted to be more stable than [Ru(H)2(H2)2(PCy3)2] (I-calc) by
DG=À25.6 kJmolÀ1. This complex does not feature any signifi-
cant interaction between the remote Lewis base NHMe and
boron with a B···N distance of 4.257 Š. Isomerization of the
CH2ÀCH2ÀNHMe chain allows to reach several other bis-s ad-
ducts associated to different conformations of the chain. In
particular, DHbend-calc, featuring a reduced B···N bond length
of 3.69 Š is computed to be more stable than that of DH-calc
by DG=À2.9 kJmolÀ1 (Figure 4). From DHbend-calc, formation
of the NÀB bond with the remote Lewis base is effective
through TS-BN-calc with an activation barrier of DG° =
71.7 kJmolÀ1. The B···N bond length is 2.03 Š in the transition-
state structure and the geometry around the boron atom has
been altered to accommodate the new bond, as illustrated by
the decrease of the Ru-B-N angle from 175.98 in the bis-s
adduct DHbend-calc to 141.78 in TS-BN-calc. The product of
the cyclization, Prod-BN-calc, lies at DG=64.7 kJmolÀ1 above
that of DHbend-calc and presents a newly formed BÀN bond
of 1.752 Š, whereas the already existing BÀN bond has been
elongated to 1.499 Š. This intermediate can be described as an
h2-Shimoi-type bis(s-amine–borane) complex with one hydro-
gen atom on one nitrogen.[19] This protic hydrogen could be
engaged in stabilizing interaction with a negatively charged
hydride. Indeed, very easy rotation around the Ru···B axis of
the 5-membered ring generates, through TS-Rot-calc, an inter-
mediate (Prod-Rot-calc), featuring a dihydrogen bond be-
tween the protic NÀH and one hydridic RuÀH (H···H=1.805 Š).
This transformation has a very low activation barrier of
DG° =2 kJmolÀ1 and is slightly exoergic (DG=À3.4 kJmolÀ1).
Interestingly, TS-Rot-calc not only sets the scene to create a di-
hydrogen bond, but, in addition, the rotation is accompanied
with the cleavage of one BÀH bond leading to Prod-Rot-calc.
that exhibits an elongated BÀH bond (BÀH=1.388 Š) and two
BÀN bonds with a distinct length (1.669 Š vs. 1.450 Š). Prod-
Rot-calc can be formulated as a borenium ruthenium com-
plex[20] in which a dihydrogen bond is established with the hy-
dride that was initially bonded to boron. From Prod-Rot-calc,
a proton transfer from the nitrogen atom to the hydride is ef-
fective through TS-NH-calc with an activation barrier of DG° =
18.5 kJmolÀ1. In the TS, the H···H bond length is reduced to
1.105 Š, whereas the NÀH bond length is elongated to 1.251 Š
(N-H=1.029 Š in Prod-Rot-calc). This NÀH bond cleavage is ac-
Table 4. Selected geometrical parameters.[a]
RuÀB
BÀN
%p LP(N)
DG°
D-calc
TSrot-D-calc
19-calc
TSrot-19-calc
DH-calc
TSrot-DH-calc
A1-calc
TSrot-A1-calc
A2-calc
TSrot-A2-calc
A3-calc
1.972
1.908
1.971
1.904
1.974
1.905
1.968
1.904
1.967
1.902
1.957
1.895
1.407
1.424
1.407
1.426
1.404
1.423
1.407
1.431
1.404
1.432
1.406
1.436
100
100
100
99
100
99.2
100
97
100
94.7
100
91.3
59.6
54.7
55.8
50.7
39.5
35.0
TSrot-A3-calc
[a] Bond lengths [Š] for the bis-s adducts and the transition state for ro-
tation around the BÀN bond, %p character in the lone pair on the nitro-
gen bonded to boron from NBO analysis, and Gibbs free energy DG°
[kJmolÀ1] of the transition state for rotation around the BÀN bond rela-
tive to the bis-s adduct.
The calculated activation barriers DG° for the rotation
around the BÀN bond for D-calc and 19-calc (59.6 and
54.7 kJmolÀ1, respectively) are in excellent agreement with the
experimental values (Table 2). This lends further credit to the
interpretation of the observed exchange process resulting
from a rotation around the BÀN bond. Calculations on DH-
calc, an analogue of D-calc with a remote NHMe group instead
of NMe2, yielded results very similar with an activation barrier
of DG° =55.8 kJmolÀ1. In fact, the activation barrier of the ro-
tation around the BÀN bond is essentially affected by the
nature of the substituents bonded to nitrogen as illustrated by
the results for the various bis-s adducts of H2BNHnÀ1Me3Àn (A1-
calc, n=1; A2-calc, n=2; A3-calc, n=3; see Table 4). When
one hydrogen is bonded to N (A2-calc and A3-calc), the acti-
vation barrier for the rotation is significantly reduced and the
hybridization at nitrogen changes in the transition state, with
more s character in the lone pair resulting in a more pyrami-
dalized nitrogen (Table 4). The value of the rotational barrier,
below 40 kJmolÀ1, is consistent with fast exchange at room
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13085
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Figure 4. Gibbs free energy profile (kJmol<-M>1, 298 K) along the pathway for the formation of the cyclic borane from the bis-s adduct.
companied by a shortening of the B···N distance that now be-
comes closer to the other one (1.579 and 1.431 Š in TS-Rot-
calc vs. 1.669 Š and 1.450 Š in Prod-BN-calc, respectively). The
monoborane substrates, with more sterically encumbered or
more rigid backbones causing the CDC process to take more
time to reach complete conversion. Exchanging the nature of
the peripheral amine unit by a methoxy group or from secon-
dary to tertiary, however, produces solely a stoichiometric reac-
tion. The corresponding complexes 19 and 20 were isolated
without any detectable interaction between the peripheral
Lewis base moiety and the ruthenium core, neither in the solid
state nor in solution. However, characterization of the postulat-
ed bis-s-BÀH bound ruthenium aminoborane catalytic inter-
mediate with a reactive NHMe end group was successful by
lengthening the carbon chain of the starting diamine–mono-
borane to six carbon atoms. Using theoretical calculations, and
starting from the key intermediate (the bis-s-BÀH aminobor-
ane ruthenium possessing an NHMe dangling moiety), we can
propose a cyclization pathway that operates by a stepwise
mechanism, resulting from sequential BÀH and NÀH bond acti-
vation. The mechanism involves the successive formation of
1) a h2-Shimoi-type bis(s-amine–borane) complex with a newly
formed BÀN bond; 2) The stabilization of this complex by BÀH
bond activation into a s-BÀH borenium ruthenium complex
displaying a RuÀH···HÀN dihydrogen interaction; Then, 3) NÀH
bond cleavage leading to the corresponding diaminoborane li-
gated in a s-BÀH fashion to the ruthenium and further dis-
placed by the produced dihydrogen, thus regenerating the
starting bis(dihydrogen) catalyst (I).
product
of
the
reaction,
Prod-NH-calc,
features
a [Ru(H)2(H2)(PCy3)2] moiety with an h2-BH coordination of the
product of the DC reaction.
Figure 4 shows the Gibbs free energy profile along the cycli-
zation pathway for the formation of cyclic borane from the
bis-s adduct. This transformation is computed to be strongly
exergonic and the substitution of the h2-BH bond in Prod-NH-
calc by a dihydrogen ligand is also energetically favored. The
rate-determining step is the NÀH cleavage step regenerating
a H2 ligand in the coordination sphere of the metal. From the
computed values in Figure 4, formation of the bis-s adduct
DHbend-calc from the catalyst I-calc and the free amine–
borane 1-calc is computed to be favored energetically.
However, the cyclization process from DHbend-calc is very
easy, produces the cyclic-borane 9-calc, and regenerates the
catalyst in a strongly exergonic transformation. The activation
barrier corresponds to a half-life of the bis-s adduct of about
10 seconds at 298 K. This is in perfect agreement with the ex-
perimental observation in which no bis-s adduct intermediate
could be observed during the CDC reaction, unless the chain is
made sufficiently long to retard the BÀN bond formation. Also,
the h2-BH coordination of the cyclic borane is not energetically
favored over H2 coordination, thus explaining why the bis-dihy-
drogen complex is the resting state of the catalyst.
Investigations with different metal complexes are currently
underway to further explore the reactivity of diamine–mono-
boranes in dehydrogenative processes.
Conclusion
We have demonstrated that the [RuH2(h2-H2)2(PCy3)2] complex
(I) is an excellent and versatile catalyst for the CDC transforma-
tion. It is capable of converting a range of different diamine–
monoborane substrates to their cyclic diaminoborane products
through a clean process at room temperature. We have also
showed that this reaction could be extended to an amine–
borane alcohol. The CDC transformation is affected by the
steric bulk and the rigidity of the backbone of the diamine–
Experimental Section
General methods
All experiments were performed under an atmosphere of dry
argon using standard Schlenk and glovebox techniques. Unless
stated, all chemicals were purchased from Aldrich and used with-
out further purification. N,N’-Di-tert-butylethylenediamine, N,N’-di-
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13086
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
benzylethylenediamine, N,N’-dimethylhexanediamine, and N,N,N’-
trimethylpropanediamine were dried over calcium hydride and de-
gassed (freeze–pump–thaw) prior to use. [RuH2(h2-H2)2(PCy3)2] was
prepared according to a literature procedure.[21] All other solvents
were purified and dried through an activated alumina purification
system (MBraun SPS-800). NMR solvents were dried using appropri-
ate methods and degassed prior to use. NMR samples of sensitive
compounds were prepared under an argon atmosphere. Nuclear
magnetic resonance spectra were recorded on Bruker AV300, 400,
or 500 spectrometers operating at 300.130, 400.130, or
500.330 MHz, respectively, for 1H; 121.495, 161.976, or
202.537 MHz, respectively, for 31P; 75.468, 100.613, or 125.808 MHz,
respectively, for 13C; 96.293, 128.377, 160.526 MHz, respectively, for
11B; and 30.420, 40.560 or 50.712 MHz for 15N. 1H and 13C NMR
chemical shifts are reported in ppm referenced internally to residu-
al protio-solvent, whereas 31P, 11B, and 15N NMR chemical shifts
were referenced to external 85% H3PO4, BF3·OEt2, and MeNO2, re-
spectively. Chemical shifts are quoted in d (ppm) and coupling
constants in Hertz. The following abbreviations are used: br,
broad; s, singlet; d, doublet; t, triplet; m, multiplet; y, pseudo;
app., apparent. Elemental analyses were performed by the “in
house” service of the Laboratoire de Chimie de Coordination, Tou-
louse. Infrared spectra were recorded on a PerkinElmer Spectrum
100 FTIR spectrometer fitted with ATR accessories. The crystal in
the ATR module is Ge. Purifications by UPLC were performed by
the HPLC service from the ICT (Institut de Chimie de Toulouse)
with ultra-performance liquid chromatography (UPLC) system from
Waters with a Macherey–Nagel C18 column and a UV detector.
High-resolution mass spectra were obtained at the ICT Mass Spec-
troscopy Service using a Waters Xevo G2 QTof spectrometer using
electrospray.
1H, N(BH3)H), 3.01 (app. qd, 1H, JHH =12.0, 4.17 Hz, CH2N(H)BH3),
2.39 (dddd, 1H, JHH =11.2, 8.7, 3.9, 2.0 Hz, CH2N(H)BH3), 2.24 (brq,
3H, BH3), 2.19 (dddd, 1H, JHH =12.4, 5.9, 4.0, 2.1 Hz, CH2NH), 2.09
(app. td, 1H, JHH =11.6, 4.3 Hz, CH2NH), 1.08 (s, 9H, NC(CH3)3(H)BH3),
0.86 ppm (m, 1H, N(H)C(CH3)3); 1H{11B} NMR (C6D6, 400.130 MHz,
298 K): d=2.35 ppm (brs, 3H, BH3); 13C{1H} NMR (C6D6,
100.613 MHz, 298 K): d=57.10 (s, NC(CH3)3(H)BH3), 50.46 (s,
CH2N(H)BH3), 50.38 (s, N(H)C(CH3)3), 39.02 (s, CH2NH), 29.54 (s,
NC(CH3)3(H)BH3), 26.92 ppm (s, N(H)C(CH3)3); 11B NMR (C6D6,
128.377 MHz, 298 K): d=À19 ppm (brq, 1JBH =97 Hz).
Diamine–monoborane 5: BH3·SMe2 (0.4 mL, 1 equiv, 4.2 mmol)
was added by syringe at À788C to a solution of the N,N’-dibenzyle-
thylenediamine (1 mL, 1 equiv, 4.2 mmol) in THF (5 mL). A white
precipitate immediately appeared and remained upon warming
the reaction mixture to room temperature over 1 h. The reaction
mixture was filtered and the white powder dried under vacuum to
yield the bis adduct [CH2NHBz(BH3)]2 (54 mg, 10%). The filtrate was
pumped to dryness to produce colorless oil. The oil was taken up
in toluene (4 mL) and stirred at À408C for 1 h to remove the dia-
mine starting material. The toluene solution was decanted and
then washed with pentane (2”4 mL) at À408C to yield a colorless
1
oil of 5 (212 mg, 39%). 5: H{11B} ([D8]THF, 500.330 MHz, 298 K): d=
7.45–7.10 (m, 10H, CH Ar), 5.05 (brs, 1H, N(BH3)H), 4.12 (dd, 1H,
2JHH =13.6, 3JHH =3.1 Hz, PhCHHN(H)BH3), 3.62 (dd, 1H, 2JHH =13.6,
3
3JHH =9.6 Hz, PhCHHN(H)BH3), 3.49 (d, 2H, JHH =5.0 Hz, PhCH2NH),
3.00 (m, 1H, CHHN(H)Bn), 2.72 (m, 1H, CHHN(H)BH3), 2.61 (m, 1H,
CHHN(H)Bn), 2.55 (m, 1H, CHHN(H)Bz),1.75 (1H, NHBz), 1.65 ppm
(brs, 3H, BH3); 13C{1H} NMR ([D8]THF, 125.808 MHz, 298 K): d=
141.82 (s, ipso-C PhCH2NH), 136.53 (s, ipso-C PhCH2N(H)BH3), 130.79
(s, o-CH NHCH2Ph(BH3)), 129.54 (s, m-CH NHCH2Ph(BH3)), 129.05 (s,
p-CH NHCH2Ph(BH3)), 129.03 (s, m-CH NHCH2Ph), 128.91 (s, o-CH
NHCH2Ph),127.58 (s, p-CH Ar NHCH2Ph), 60.67 (s, PhCH2N(H)BH3),
54.03 (s, PhCH2NH), 53.03 (s, CH2N(H)BzBH3), 45.66 ppm (s,
CH2NHBz); 11B NMR ([D8]THF, 128.4 MHz, 298 K): d=À15 ppm (brq,
1JBH =93 Hz). Because of the oily nature of 5 at room temperature,
no elemental analysis could be properly performed.
X-ray crystallographic studies
Data for compound 19 and 20 were collected at low temperature
(100 K) on a Gemini Agilent diffractometer using a graphite-mono-
chromated MoKa radiation (l=0.71073 Š) and equipped with an
Oxford Instrument Cooler Device. The final unit cell parameters
were obtained by means of a least-squares refinement. The struc-
tures have been solved by Direct Methods using SIR92,[22] and re-
fined by means of least-squares procedures on a F2 with the aid of
the program SHELXL97[23] included in the software package WinGX
version 1.63.[24] The atomic scattering factors were taken from In-
ternational tables for X-ray Crystallography.[25] All hydrogen atoms
were placed geometrically, and refined by using a riding model,
except for the hydrides Hy1, Hy2, Hy3, and Hy4, for both com-
pounds that were located by Fourier differences and isotropically
refined. All non-hydrogens atoms were anisotropically refined, and
in the last cycles of refinement a weighting scheme was used, in
which weights are calculated from the following formula: w=1/
[s2(Fo2)+(aP)2+bP] in which P=(Fo2+2Fc2)/3. Drawing of mole-
cule were performed with the program ORTEP32[26] with 30%
probability displacement ellipsoids for non-hydrogen atoms.
Diamine–monoborane
6 and 7: BH3·SMe2 (0.33 mL, 1 equiv,
3.5 mmol) was added by syringe at À788C to a solution of either
enantiopure R,R-(À)- or S,S-(+)-N,N’-dimethyl-1,2-cyclohexanedia-
mine (0.5 g, 1 equiv, 3.5 mmol) in THF (4 mL). The reaction solution
was allowed to warm to room temperature over 2 h. The reaction
mixture was pumped to dryness, which yielded a yellowish oil of 6
(0.336 g, 62%) or 7, respectively. The product is present in the
form of two spectroscopically different diastereoisomers, (ratio of
1
isomersꢀ90:10). Major isomer: H NMR (C6D6, 500.330 MHz, 298 K):
3
d=5.55 (br, 1H, NH(BH3)), 2.61 (app. dt, 1H, 3JHH =4.2 Hz, JHH
=
11.0 Hz, CHNHMe), 2.11 (s, 3H, CH3NH), 2.10 (d, 3H, 3JHH =6.0 Hz,
CH3NH(BH3)), 1.90 (m, 1H, CHH), 1.73 (m, 1H, CHH), 1.61 (m, 1H,
CHH), 1.58 (m, 1H, CHH), 1.50 (m, 1H, CHH),1.39 (m, 2H,
3
CHNHMe(BH3)+CHH), 0.94 (app. qt, 1H, 3JHH =13.2 Hz, JHH
=
3.6 Hz, CHH), 0.76 (app. qt, 1H, 3JHH =13.1, 3.4 Hz, CHHCHNHMe),
0.27 ppm (m, 1H, CHHCHNHMe); TOCSY 1H{11B} NMR (C6D6,
500.330 MHz, 298 K): d=2.25 (s, 3H, BH3), À0.5 ppm (br, 1H, NH);
13C{1H} NMR
(C6D6,
125.808 MHz,
298 K):
d=68.31
(s,
Syntheses
CHNHMe(BH3)), 58.37 (s, CHNHMe), 41.85 (s, CH3N(H)BH3), 33.99 (s,
Diamine–monoborane 4: BH3·SMe2 (0.55 mL, 1 equiv, 5.8 mmol)
through a syringe was added at À788C to a solution of the N,N’-
di-tert-butylethylenediamine (1.25 mL, 1 equiv, 5.8 mmol) in THF
(5 mL). A white precipitate immediately appeared and remained
upon warming the reaction mixture to room temperature over 1 h.
The reaction mixture was pumped to dryness and the resulting
white powder was dried under dynamic vacuum to yield 4
(0.77 mg, 71%). 1H NMR (C6D6, 400.130 MHz, 298 K): d=4.27 (brs,
CH3NH), 31.56 (s, CH2), 25.56 (s, CH2), 25.33 (s, CH3), 24.27 ppm (s,
1
CH2); 11B NMR (C6D6, 128.377 MHz, 298 K): d=À18.4 ppm (q, JBH
=
97 Hz); Minor isomer 13C{1H} NMR (C6D6, 125.808 MHz, 298 K): d=
66.37 (s, CHNHMe(BH3)), 58.30 (s, CHNHMe), 33.99 (s, NHCH3(BH3)),
33.08 (s, CH3), 31.03 (s, CH2), 24.7 (s, CH2), 24.1 (s, CH2), 23.92 ppm
1
(s, CH2); 11B NMR (C6D6, 128.377 MHz, 298 K): d=À14 ppm (q, JBH
=
97 Hz); Because of the oily nature of 6 and 7 at room temperature,
no elemental analysis could be properly performed.
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13087
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Pyrrolidinemethanol borane adduct 8: A 1m solution of BH3·THF
in THF (1.8 mL, 2 equiv, 1.6 mmol) was added through a syringe to
a solution of (S)-(À)-a,a-diphenyl-2-pyrrolidinemethanol (200 mg,
0.79 mmol) in THF (10 mL) at À408C. The reaction mixture was
stirred at À408C for 15 min before H2O (0.32 mL) was added to
quench the excess BH3. The mixture was evaporated to dryness
and the residue taken up in toluene (10 mL). The toluene solution
was dried with MgSO4 and extracted by filtration and washing with
toluene (5 mL). The toluene solution was reduced to a volume of
approximately 3 mL, and then placed at À188C overnight. Com-
pound 8 precipitated as a single diastereoisomer,[10] affording
a white powder that was isolated and dried under vacuum to yield
an analytically pure compound (61.5 mg, 29%). 1H NMR ([D8]THF,
Diazaborolidine 14 and 15: [RuH2(h2-H2)2(PCy3)2] (2.5 mol%;
40 mg, 0.06 mmol) in THF (5 mL) at room temperature was added
to a solution of either 6 or 7 (370 mg, 0.24 mmol) in THF (5 mL).
The reaction solution was left to stir at room temperature for 30 h.
The volatiles were distilled off by a trap-to-trap distillation under
dynamic vacuum. The THF was then separated from the diamino-
borane product by
a trap-to-trap distillation under reduced
vacuum (58C, 5 millitor) to yield a colorless oil of 14 or 15
(0.28 mg, 78%). For 14:½aŠ20 =À354Æ9 (c=2.76, CH2Cl2); for 15:
D
1
½aŠ20 = +352Æ9 (c=2.83, CH2Cl2). H{11B} NMR (C6D6, 400.130 MHz,
D
298 K): d=4.24 (br, 1H, BH), 2.62 (s, 6H, CH3N), 2.48 (br, 2H, CH),
1.86 (br, 2H, CH2), 1.59 (br, 2H, CH2), 1.08 ppm (br, 4H, CH2);
13C{1H} NMR (C6D6, 100.625 MHz, 298 K): d=69.60 (s, CHN), 32.71 (s,
3
400.130 MHz, 298 K): d=7.62 (d, 2H, JHH =7.2 Hz, CH Ph), 7.46 d,
CH3), 30.01 (s, CH2), 25.59 ppm (s, NCH3); 11B NMR (C6D6,
2H, 3JHH =7.8 Hz, CH Ph), 7.26 (app. q, 4H, Japp. =7.5 Hz, CH Ph),
7.16 (app. q, 2H, Japp. =7.1 Hz, CH Ph), 4.71 (br, 1H, NH), 4.35 (br,
1H, NCH), 3.16 (m, 1H, NCHH), 2.88 (br, 1H, NCHH), 2.14 (m, 1H,
NCHCHH), 2.12 (m, 1H, NCH2CHH), 1.80 (m, 1H, NCH2CHH),
128.377 MHz, 298 K): d=31.7 ppm (d, JBH =142 Hz); IR (neat): n=
1
~
2541 and 2507 cmÀ1 (w, BH); HRMS ESI+: m/z calcd for C8H18BN2
:
+
152.1599 [M+H+]; found: 152.1595 (2.6 ppm), exact agreement be-
tween the experimental and theoretical isotopic peak distributions,
the accurate mass is measured and calculated on the mono-isotop-
ic peak.
1
1.69 ppm (m, 1H, NCHCHH); H{11B} NMR (C6D6, 400.1 MHz, 298 K):
d=2.46 ppm (br, 3H, BH3). 13C{1H} NMR (C6D6, 100.612 MHz, 298 K):
d=144.88, 144.32 (s, ipso-C Ph), 129.18, 128.94, 127.83, 127.01,
126.60 (s, CH Ph), 80.78 (s, Ph2COH), 72.36 (s, NCH), 55.33 (s, NCH2),
27.54 (s, NCH2CH2), 24.55 ppm (s, NCHCH2); 11B NMR (C6D6,
Oxazaborolidine 16: [RuH2(h2-H2)2(PCy3)2] (2.5 mol%; 3.8 mg,
0.05 mmol) in THF (3 mL) at room temperature was added to a solu-
tion of the (S)-(À)-a,a-diphenyl-2-pyrrolidinemethanol–monobor-
ane 8 (58.1 mg, 0.02 mmol) in THF (3 mL). The reaction solution
was left to stir at room temperature for 24 h, before the solvent
was removed under vacuum and the reaction mixture taken up in
toluene (1 mL). Pentane (3 mL) was added and a white powder
precipitated. Filtration, then washing with pentane (1 mL), followed
by drying under vacuum yielded a white powder analyzed as oxa-
zaborolidine 16 (24.0 mg, 42%), as previously reported in the liter-
ature.[11] Selected data: 11B{1H} NMR ([D8]THF, 128.4 MHz, 298 K): d=
28.4 (brs), 7.4 ppm (brs).
1
128.377 MHz, 298 K): d=À12.4 ppm (br, JBH =81 Hz, BH3); elemen-
tal analysis calcd (%) for C17H22BNO: C 76.42, H 8.30, N 5.24; found:
C 76.46, H 8.60, N 4.97.
N,N’-Di-tert-butyl-1,3,2-diazaborolidine 12: [RuH2(h2-H2)2(PCy3)2]
(2.5 mol%) (14 mg, 0.04 mmol) in THF (4 mL) at room temperature
was added to a solution of 4 (0.3 mg, 1.6 mmol) in THF (4 mL). The
reaction solution was left to stir at room temperature for 16 h. The
volatiles were distilled off through a trap-to-trap distillation under
dynamic vacuum. The THF was then separated from the diamino-
borane product by
a trap-to-trap distillation under reduced
N-(2-Methoxyethyl)methylamine borane adduct 17: BH3·SMe2
(1 mL, 1 equiv, 10.54 mmol) was added by syringe at room temper-
ature to an ethereal solution of N-(2-methoxyethyl)methylamine
(1.132 mL, 10.54 mmol). The reaction solution was left to stir at
room temperature for 24 h. The mixture was evaporated to dryness
vacuum (258C, 5 millitor) to yield a colorless oil of 12 (0.26 mg,
1
88%). H{11B} NMR (C6D6, 400.130 MHz, 298 K): d=4.42 (s, 1H, BH),
3.13 (s, 4H, CH2N), 1.18 ppm (s, 18H, C(CH3)3); 13C{1H} NMR (C6D6,
100.6 MHz, 298 K): d=50.60 (s, C(CH3)3), 45.28 (s, CH2N), 30.59 ppm
(s, C(CH3)3); 11B NMR (C6D6, 128.377 MHz, 298 K)À: 1d=26.5 ppm (d,
to yield 17 quantitatively as colorless oil. 1H NMR (C6D6,
1JBH =140 Hz); IR (neat): n=2580 and 2543 cm (w, BH); HRMS
3
~
300.130 MHz, 298 K): d=3.51 (brs, 1H, NH), 3.25 (ddd, 1H, JHH
=
+
ESI+: m/z calcd for C10H24BN2
:
182.2069 [M+H+]; found:
10.7, 7.5, 3.5 Hz, 1H, MeNCHH), 2.89 (s, 3H, OCH3), 2.80 (br, 1H,
MeNCHH), 2.56 (m, 1H, CHHOMe), 2.21 (brq, 3H, BH3), 2.12 (m, 1H,
182.2072 (1.6 ppm); exact agreement between the experimental
and theoretical isotopic peak distributions, the accurate mass is
measured and calculated on the mono-isotopic peak.
3
CHHOMe), 1.92 ppm (d, 3H, JHH =4.6 Hz, NCH3); 13C{1H} NMR (C6D6,
75.468 MHz, 298 K): d=67.54 (s, OCH2), 58.76 (s, OCH3), 56.54 (s,
CH2N), 42.43 ppm (s, NCH3); 11B NMR (C6D6, 96.293 MHz, 298 K): d=
À14 ppm (q, 1JBH =98 Hz); Because of the oily nature of 17 at
room temperature, no elemental analysis could be properly per-
formed.
N,N’-Dibenzyl-1,3,2-diazaborolidine
13:
[RuH2(h2-H2)2(PCy3)2]
(2.5 mol%; 14 mg, 0.03 mmol) in THF (3 mL) at room temperature
was added to a solution of 5 (212 mg, 0.83 mmol) in THF (3 mL).
The reaction solution was left to stir at room temperature for 72 h,
before the solvent was removed under vacuum and the reaction
mixture taken up in pentane (4 mL). After filtration and washing
with pentane (4 mL), the pentane fraction was reduced to approxi-
mately 1 mL and placed at À358C overnight. The pentane solution
was then decanted off and the resulting oil dried under vacuum to
Diamine–monoborane adduct 18: A 1m solution of BH3·THF in
THF (3.4 mL, 1 equiv, 3.4 mmol) was added by syringe at room
temperature to a neat solution of N,N,N’-trimethylpropanediamine
(2 mL, 4 equiv, 14 mmol). The reaction solution was left to stir at
room temperature for 24 h. The mixture was evaporated to dryness
and re-dissolved in CH2Cl2 (10 mL) and passed through a small
plug of silica gel. The silica was washed with CH2Cl2 (5 mL) and the
combined CH2Cl2 fractions were pumped to dryness to yield a col-
orless oil of 18 (193 mg, 43.5%). 1H{11B} NMR (C6D6, 400.130 MHz,
298 K): d=5.63 (brs, 1H, NH), 2.73 (m, 1H, MeN(H)CHH), 2.37 (d,
1
yield 13 (151 mg, 72%). H NMR ([D8]THF, 400.130 MHz, 298 K): d=
7.33–7.22 (m, 10H, Ar-H), 4.15 (s, 4H, PhCH2N), 4.08 (br, 3H, BH),
3.07 ppm (s, 4H, CH2N); 13C{1H} NMR ([D8]THF, 100.613 MHz, 298 K):
d=141.60 (s, ipso-C Ph), 129.16 (s, m-CH Ph), 128.46 (s, o-CH Ph),
127.59 (s, p-CH Ph), 53.03 (s, PhCH2N), 49.58 ppm (s, NCH2CH2N);
11B NMR ([D8]THF, 128.377 MHz, 298 K): d=29.4 ppm (q, JBH
1
3
3
=
3H, JHH =3.1 Hz, BH3), 2.07 (d, 3H, JHH =6.0 Hz, NHCH3(BH3)), 2.06
À1
3
~
129 Hz); IR (neat): n=2553 and 2527 cm (w, BH); HRMS ESI+:
m/z calcd for C16H20BN2+: 250.1756 [M+H+]; found: 250.1759
(1.2 ppm), exact agreement between the experimental and theo-
retical isotopic peak distributions, the accurate mass is measured
and calculated on the mono-isotopic peak.
(m, 1H, NCHH), 1.90 (dddd, 1H, JHH =12.5, 6.9, 3.7, 0.8 Hz, NCHH),
1.83 (s, 6H, N(CH3)2), 1.80 (m, 1H, NCHH) 1.60 (m, 1H, NCH2CHH),
0.98 ppm (dtt, 1H, 3JHH =15.2, 7.1, 3.6 Hz, NCH2CHH); 13C{1H} NMR
(C6D6, 100.613 MHz, 298 K): d=60.27 (s, NCH2), 58.92 (s, NCH2),
45.53 (s, N(CH3)2), 42.61 (s, CH3) 22.44 ppm (s, NCH2CH2); 11B NMR
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13088
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
3
(C6D6, 128.377 MHz, 298 K): d=À14 ppm (q, 1JBH =96.5 Hz). Be-
cause of the oily nature of 18 at room temperature, no elemental
analysis could be properly performed.
2.37 (s, 3H, NHCH3), 2.12 (br, 1H, NHCH3), 1.60 (p, 2H, JHH =7.6 Hz,
3
CH2CH2NHMe(BH3)), 1.45 (p, 2H, JHH =7.1 Hz, CH2CH2NHMe), 1.36–
1.18 ppm (m, 4H, CH2 +CH2); 1H{11B} NMR ([D8]THF, 500.330 MHz,
298 K): d=1.48 ppm (br, 3H, BH3); 13C{1H} NMR (CDCl3, 75.468 MHz,
298 K): d=56.69 (s, CH2NHMe(BH3)), 51.66 (s, CH2NHMe), 41.79 (s,
Complex 19:
A
solution of [RuH2(h2-H2)2(PCy3)2] (100 mg,
0.01 mmol) in toluene (5 mL) at room temperature was aaded to
a solution of 17 (15 mg, 1.5 equiv, 0.15 mmol) in toluene (2 mL).
The reaction was allowed to stir at room temperature for 2 h, in
which time the reaction solution turned orange. The reaction mix-
ture was pumped to dryness to give a pale-yellow solid. The solid
was extracted with pentane (2”8 mL) and after filtration the pen-
tane filtrant was reduced to half volume and placed at À358C. The
pale-yellow crystalline precipitate of the desired compound was
isolated by filtration and dried under vacuum to yield 19 (119 mg,
83%). Pale yellow, almost colorless crystals suitable for X-ray analy-
NH(CH3)(BH3)), 36.27 (s, N(CH3)H), 29.29, 26.68, 26.53, 26.18 ppm (s,
CH2); 11B NMR (C6D6, 96.293 MHz, 298 K): d=À15 ppm (q, JBH
=
=
1
99 Hz); 15N NMR ([D8]THF, 50.717 MHz, 273 K): d=À351 (d, J15NH
1
67 Hz, N(H)BH3), À357 ppm (br, d, NH). Because of the waxy nature
of 21 at room temperature, no elemental analysis could be proper-
ly performed.
N-Methylhexylamine borane adduct 25:
A 1m solution of
BH3·THF in THF (13.2 mL, 1 equiv, 13.19 mmol) was added by sy-
ringe to an ethereal solution (5 mL) of N-methylhexylamine (2 mL,
1.52 g, 13.19 mmol) at À208C. The reaction solution is left to stir at
room temperature for 2 h and the mixture is evaporated to dryness
to yield quantitatively 25 as colorless oil. 1H{11B} NMR ([D8]THF,
400.130 MHz, 298 K): d=4.82 (brs, 1H, NH), 2.71–2.60 (m, 1H,
MeNCHH), 2.74–2.59 (m, 1H, MeNCHH), 2.36 (d, 3H, 3JHH =5.6 Hz,
NHCH3), 1.70–1.20 (m, 4”CH2), 1.30 (brs, 3H, BH3), 0.89 ppm (t, 3H,
3JHH =6.20 Hz, CH3 hex); 13C{1H} NMR ([D8]THF, 100.613 MHz, 298 K):
d=57.79 (s, NCH2), 42.35 (s, NCH3), 32.62, 27.70, 27.42, 23.55 (s,
CH2), 14.52 ppm (s, CH3 hex); 11B NMR ([D8]THF, 128.378 MHz,
298 K): d=À14.5 ppm (q, 1JBH =95 Hz); 15N NMR ([D8]THF,
1
sis were grown by slow evaporation of a pentane solution. H NMR
3
([D8]Tol, 400.1 MHz, 298 K): d=3.38 (t, 2H, JHH =5.9 Hz, NCH2), 3.23
3
(t, 2H, JHH =6.0 Hz, NCH2), 3.15 (s, 3H, OCH3), 2.83 (s, 3H, NCH3),
2.29–1.12 (m, 66H, Cy), À7.05 (br, 2H, s-BH2), À12.35 ppm (br, 2H,
RuH2); 13C{1H} NMR ([D8]Tol, 100.613 MHz, 298 K): d=72.73 (s,
OCH2), 58.92 (s, OCH3), 52.51 (s, NCH2), 39.49 (yt, CH, Cy), 38.43 (s,
NCH3), 31.43(s, CH2 Cy), 28.90 (yt, CH2 Cy), 27.69 ppm (s, CH2 Cy);
31P{1H} NMR ([D8]Tol, 161.975 MHz, 298 K): d=79.2 ppm (s);
11B{1H} NMR ([D8]Tol, 128.377 MHz, 298 K): d=47.6 ppm (br); ele-
mental analysis calcd (%) for C40H80BNOP2Ru: C 62.81, H 10.54, N
1.83; found: C 62.90, H 10.99, N, 1.98.
1
40.560 MHz, 298 K): d=À350 ppm (d, JNH =71 Hz). Because of the
oily nature of 25 at room temperature, no elemental analysis could
Complex 20: A solution of the diamine–monoborane 18 (64 mg,
1.5 equiv, 0.49 mmol) in toluene (4 mL) was added to a solution of
[RuH2(h2-H2)2(PCy3)2] (220 mg, 0.33 mmol) in toluene (10 mL) at
room temperature. The reaction was allowed to stir at room tem-
perature for 2 h, in which time the reaction solution turned
orange. The reaction mixture was pumped to dryness to give
a pale-yellow solid. The solid was extracted with pentane (2”8 mL)
and after filtration the pentane filtrant was reduced to half volume
and placed at À358C. The pale-yellow crystalline precipitate of the
desired compound was isolated by filtration and dried under
vacuum to yield 20 (119 mg, 46%). Pale yellow, almost colorless
crystals suitable for X-ray analysis were grown by slow evaporation
be properly performed.
Complex 26: A solution of 25 (21 mg, 0.15 mmol) in toluene
(5 mL) was added to a solution of [RuH2(h2-H2)2(PCy3)2] (100 mg,
0.15 mmol) in toluene (10 mL) at room temperature. The reaction
was allowed to stir at room temperature for 1 h, in which time the
reaction solution turns yellow. The toluene was then removed
under vacuum, to give a pinkish solid. The solid was taken up in
pentane (1 mL) and placed at À358C overnight. An off-white solid
precipitated overnight, and after isolating and drying under
vacuum, yielded analytically pure 26 (50 mg, 42%). 1H NMR
([D8]Tol, 400.130 MHz, 298 K): d=3.06 (t, 2H, 3JHH =7.4 Hz,
BN(CH3)CH2), 2.76 (s, 3H, NCH3), 2.13–2.24 (m, 12H, Cy), 2.00–1.10
(m, 62H, Cy+4”CH2 hex), 0.91 (t, 3H,3JHH =6.5 Hz, CH3 hex), À7.06
(br, 2H, RuH2B), À12.38 ppm (br, 2H, RuH2); 13C{1H} NMR ([D8]Tol,
100.613 MHz, 298 K): d=53.22 (s, NCH2), 39.51 (yt, CH, Cy), 37.23
(s, NCH3), 32.79 (s, CH2 hex), 31.45 (s, CH2 Cy), 29.50 (s, CH2 hex),
28.91 (yt, CH2 Cy), 27.90 (s, CH2 Cy), 27.61 (s, CH2 hex), 23.54 (s, CH2
hex), 14.72 ppm (s, CH3 hex); 31P{1H} NMR ([D8]Tol, 161.975 MHz,
298 K): d=79.4 ppm (s); 11B{1H} NMR ([D8]Tol, 128.377 MHz, 298 K):
d=46.9 ppm (br); 15N NMR ([D8]THF, 40.560 MHz, 298 K): d=
À299 ppm (s, H2BN); elemental analysis calcd (%) for C43H86BNP2Ru:
C 65.29, H 10.96, N 1.77; found: C 65.52, H 11.62, N 1.65.
1
of a pentane solution. H NMR (C6D6, 400.130 MHz, 298 K): d=3.17
3
(t, 2H, JHH =7.4 Hz, NCH2), 2.80 (s, 3H, NCH3), 2.35–2.20 (m, 14H,
Cy), 2.15 (s, 6H, N(CH3)2), 1.95–1.20 (m, 58H, Cy+3”CH2), À6.99
(br, 2H, RuH2B), À12.27 ppm (br, 2H, RuH2); 13C{1H} NMR (C6D6,
100.613 MHz, 298 K): d=57.77 (s, NCH2), 50.88 (s, NCH2), 45.76 (s,
N(CH3)2), 39.15 (yt, Hz, CH Cy), 37.12 (s, NCH3), 31.11 (s, CH2 Cy),
28.54 (yt, CH2 Cy), 27.55 (s, NCH2CH2CH2N), 27.52 ppm (m, CH2 Cy);
31P{1H} NMR (C6D6, 161.975 MHz, 298 K): d=79.3 ppm (s);
11B{1H} NMR (C6D6, 128.377 MHz, 298 K): d=47.4 ppm (br); elemen-
tal analysis calcd (%) for C42H85BN2P2Ru: C 63.70, H 10.82, N 3.54;
found: C 63.72, H 10.19, N 3.42.
CCDC 1056987 (19) and CCDC-1056986 (20) contain the supple-
mentary crystallographic data. These data can be obtained free of
charge by The Cambridge Crystallographic Data Centre.
Diamine–monoborane 21: BH3·SMe2 (0.27 mL, 5.6 mmol) was
added by syringe to a solution of N,N’-dimethyl-1,6-hexanediamine
(1 mL, 5.6 mmol) in THF (8 mL) at À788C. The reaction solution
was allowed to warm to room temperature over 1 h. The reaction
mixture was pumped to dryness to yield a white wax as a mixture
(839 mg) of 21 and the bis-adduct of [(CH2)3NHMe(BH3)]. The mix-
ture was purified by UPLC using a H2O/MeOH (95:5, v/v) eluent
containing 0.1%v HCO2H. The product is detected by UV spectros-
copy at 210 nm. The collected fractions are gathered, neutralized
with Na2CO3 until a pH of 11 is reached, and extracted with CH2Cl2.
The organic layer is dried with MgSO4, filtered and evaporated to
dryness to give 21 as a white wax. 1H NMR (CDCl3, 300.130 MHz,
298 K): d=4.16 (brs, 1H, NH(BH3)), 2.84–2.65 (m, 1H,
CHHNHMe(BH3)), 2.47–2.60 (m, 1H, CHHNHMe(BH3)), 2.52 (t, 2H,
3JHH =7.0 Hz, CH2NHMe), 2.44 (d, 3H, 3JHH =5.7 Hz, CH3NH(BH3)),
Computational details
DFT calculations were performed with Gaussian 09 D.01,[27] with the
hybrid B3PW91 functional.[28] The Ru atom was represented by the
relativistic effective core potential (RECP) from the Stuttgart group
and the associated basis set augmented by a f polarization func-
tion (a=1.235).[29] The remaining atoms (C, H, B, N, O) were repre-
sented by a 6–31G(d,p) basis set.[30] The P atom was represented
by RECP from the Stuttgart group and the associated basis set,[31]
augmented by a d polarization function (a=0.387).[32] The solvent
(thf) influence was taken into consideration through single point
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13089
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[16] D. J. Grant, D. A. Dixon, J. Phys. Chem. A 2006, 110, 12955–12962.
[17] G. Alcaraz, S. Sabo-Etienne, Coord. Chem. Rev. 2008, 252, 2395–2409.
[18] All the optimized structures discussed in this manuscript are available
in a single xyz file in the Supporting Information, allowing the geome-
tries to be viewed.
[19] R. Dallanegra, A. B. Chaplin, A. S. Weller, Angew. Chem. Int. Ed. 2009, 48,
6875–6878; Angew. Chem. 2009, 121, 7007–7010.
[20] a) W. E. Piers, S. C. Bourke, K. D. Conroy, Angew. Chem. Int. Ed. 2005, 44,
5016–5036; Angew. Chem. 2005, 117, 5142–5163; b) T. Stahl, K. Muther,
Y. Ohki, K. Tatsumi, M. Oestreich, J. Am. Chem. Soc. 2013, 135, 10978–
10981.
calculations on the gas-phase optimized geometry within the SMD
model,[33] using the ORCA software.[34] For the solvent calculations
the pseudo potential was kept on Ru and all the remaining atoms
were treated with tzvpp basis sets.[35] The influence of the disper-
sion interactions was taken into account using the D3(bj) correc-
tion introduced by Grimme.[36] The energies reported in this work
are Gibbs free energies obtained by summing the electronic
energy obtained with the smd model augmented by the gas
phase Gibbs correction at 298 K and the D3(bj) contribution.
Table S2 in the Supporting Information collects the values of the
energy, the Gibbs correction and the dispersion corrections for all
the structures optimized.
[21] A. F. Borowski, S. Sabo-Etienne, M. L. Christ, B. Donnadieu, B. Chaudret,
Organometallics 1996, 15, 1427–1434.
[22] A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, J. Appl. Crystal-
logr. 1993, 26, 343–350.
[23] G. M. Sheldrick in SHELX97 (Includes SHELXS97, SHELXL97, CIFTAB)–
Programs for Crystal Structure Analysis (Release 97–2), Institüt für Anor-
ganische Chemie der Universität, Tammanstrasse 4, D-3400 Gçttingen,
Germany, 1998.
[24] L. J. Farrugia, J. Appl. Crystallogr. 1999, 32, 837–838.
[25] in International Tables for X-ray crystallography, Vol. IV Kynoch press,
Birmingham, England, 1974.
Acknowledgements
We thank the CNRS and the ANR for support through the ANR
program HyBoCat (ANR-09-BLAN-0184).
[26] L. J. Farrugia, J. Appl. Crystallogr. 1997, 30, 565–565.
[27] Gaussian 09, Revision D1, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, J. E. Peralta, F. Ogliaro, M.
Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Kobaya-
shi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyengar, J.
Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B. Cross, V.
Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann, O. Yazyev,
A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin, K. Moro-
kuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg, S. Dap-
prich, A. D. Daniels, Ö. Farkas, J. B. Foresman, J. V. Ortiz, J. Cioslowski,
D. J. Fox, Wallingford CT, 2009.
[28] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–5652; b) J. P. Perdew, Y.
Wang, Phys. Rev. B 1992, 46, 12947–12954.
[29] D. Andrae, U. Häußermann, M. Dolg, H. Stoll, H. Preuß, Theor. Chim.
Acta 1990, 77, 123–141.
[30] P. C. Hariharan, J. A. Pople, Theor. Chim. Acta 1973, 28, 213–222.
[31] A. Bergner, M. Dolg, W. Küchle, H. Stoll, H. Preuß, Mol. Phys. 1993, 80,
1431–1441.
[32] A. Hçllwarth, M. Bçhme, S. Dapprich, A. W. Ehlers, A. Gobbi, V. Jonas,
K. F. Kçhler, R. Stegmann, A. Veldkamp, G. Frenking, Chem. Phys. Lett.
1993, 208, 237–240.
Keywords: boranes · cyclization · dehydrogenation · density
functional calculations · ruthenium · synthetic methods
[1] a) G. Alcaraz, S. Sabo-Etienne, Angew. Chem. Int. Ed. 2010, 49, 7170–
7179; Angew. Chem. 2010, 122, 7326–7335; b) H. C. Johnson, T. N.
Hooper, A. S. Weller in The Catalytic Dehydrocoupling of Amine–Boranes
and Phosphine–Boranes, Vol. 49 (Eds.: E. Fernµndez, A. Whiting), Spring-
er, Heidelberg, 2015, pp. 153–220.
[2] A. Staubitz, A. P. M. Robertson, I. Manners, Chem. Rev. 2010, 110, 4079–
4124.
[3] a) N. C. Smythe, J. C. Gordon, Eur. J. Inorg. Chem. 2010, 509–521; b) A.
Staubitz, M. E. Sloan, A. P. M. Robertson, A. Friedrich, S. Schneider, P. J.
Gates, J. Schmedt auf der Günne, I. Manners, J. Am. Chem. Soc. 2010,
132, 13332–13345.
[4] G. Alcaraz, E. Clot, U. Helmstedt, L. Vendier, S. Sabo-Etienne, J. Am.
Chem. Soc. 2007, 129, 8704–8705.
[5] Y. Gloaguen, G. BØnac-Lestrille, L. Vendier, U. Helmstedt, E. Clot, G. Alcar-
az, S. Sabo-Etienne, Organometallics 2013, 32, 4868–4877.
[6] G. Alcaraz, L. Vendier, E. Clot, S. Sabo-Etienne, Angew. Chem. Int. Ed.
2010, 49, 918–920; Angew. Chem. 2010, 122, 930–932.
[7] G. Alcaraz, A. B. Chaplin, C. J. Stevens, E. Clot, L. Vendier, A. S. Weller, S.
Sabo-Etienne, Organometallics 2010, 29, 5591–5595.
[8] G. J. Kubas, J. Organomet. Chem. 2014, 751, 33–49.
[9] C. J. Wallis, H. Dyer, L. Vendier, G. Alcaraz, S. Sabo-Etienne, Angew. Chem.
Int. Ed. 2012, 51, 3646–3648; Angew. Chem. 2012, 124, 3706–3708.
[10] H. Tlahuext, F. Santiesteban, E. García-Bµez, R. Contreras, Tetrahedron:
Asymmetry 1994, 5, 1579–1588.
[11] E. J. Corey, R. K. Bakshi, S. Shibata, J. Am. Chem. Soc. 1987, 109, 5551–
5553.
[12] J. Graton, M. Berthelot, C. Laurence, Perkin Trans. 2 2001, 2130–2135.
[13] Traces of a bis-s-aminoborane Ru complex were detected at the end of
the reaction along with I regenerated as a result of evolution of dihy-
drogen (see the Supporting Information of ref. [9]). The bis-s ligated
aminoborane in complex D is not easily displaced by dihydrogen (see
S34–35 in the Supporting Information).
[33] A. V. Marenich, C. J. Cramer, D. G. Truhlar, J. Phys. Chem. B 2009, 113,
6378–6396.
[34] F. Neese, WIREs Comput. Mol. Sci. 2012, 2, 73–78.
[35] A. Schäfer, H. Horn, R. Ahlrichs, J. Chem. Phys. 1992, 97, 2571–2577.
[36] a) S. Grimme, S. Ehrlich, L. Goerigk, J. Comput. Chem. 2011, 32, 1456–
1465; b) S. Grimme, J. Antony, S. Ehrlich, H. Krieg, J. Chem. Phys. 2010,
132, 154104–154119.
[14] R. Custelcean, J. E. Jackson, Chem. Rev. 2001, 101, 1963–1980.
[15] A. J. Ashe, J. W. Kampf, C. Müller, M. Schneider, Organometallics 1996,
15, 387–393.
Received: April 22, 2015
Published online on July 28, 2015
Chem. Eur. J. 2015, 21, 13080 – 13090
www.chemeurj.org
13090
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim