Mechanistic Studies on Ruthenium(II)-Catalyzed Base-Free Transfer Hydrogenation Triggered by Roll-Over Cyclometalation

Article abstract of DOI:10.1002/cplu.201600526

The synthesis of 2-substituted pyridine–pyrimidine ligands and their complexation with arene ruthenium(II) chloride moieties is reported. Depending on the electronic and steric influences of the ligand, the catalysts undergo CH activation by roll-over cyclometalation. This process opens up the route to the catalytic transfer hydrogenation of ketones with isopropanol as the hydrogen source under base-free and mild conditions. Barriers related to the roll-over cyclometalation process can be determined experimentally by collision-induced dissociation ESI mass spectrometry. They are supported by DFT calculations and allow the classification of the ligands according to their electronic and steric properties, which is also in accordance with critical bond parameters derived from X-ray structure data. DFT calculations furthermore reveal that the formation of a ruthenium(II) hydrido species is plausible through β-hydride elimination from isopropanol.

Full text of DOI:10.1002/cplu.201600526

DOI: 10.1002/cplu.201600526
Full Papers
Mechanistic Studies on Ruthenium(II)-Catalyzed Base-Free
Transfer Hydrogenation Triggered by Roll-Over
Cyclometalation
[
a]
[a, b]
[a, b]
[c]
[a]
Christian Kerner, Johannes Lang,
Maximilian Gaffga,
Fabian S. Menges, Yu Sun,
[a, b]
[a]
Gereon Niedner-Schatteburg,*
and Werner R. Thiel*
The synthesis of 2-substituted pyridine–pyrimidine ligands and
their complexation with arene ruthenium(II) chloride moieties
is reported. Depending on the electronic and steric influences
of the ligand, the catalysts undergo CH activation by roll-over
cyclometalation. This process opens up the route to the cata-
lytic transfer hydrogenation of ketones with isopropanol as the
hydrogen source under base-free and mild conditions. Barriers
related to the roll-over cyclometalation process can be deter-
mined experimentally by collision-induced dissociation ESI
mass spectrometry. They are supported by DFT calculations
and allow the classification of the ligands according to their
electronic and steric properties, which is also in accordance
with critical bond parameters derived from X-ray structure
data. DFT calculations furthermore reveal that the formation of
a ruthenium(II) hydrido species is plausible through b-hydride
elimination from isopropanol.
Introduction
In 1925, Meerwein and Schmidt, Pondorf, Verley, and Oppenha-
uer independently from each other discovered the transfer hy-
In 2013, we reported that ruthenium(II) complexes with N,N’-
chelating 2-(2-aminopyrimidin-4-yl)pyridine ligands could un-
dergo a self-activation process: cleavage of the ortho-CH bond
at the pyrimidine ring through a roll-over cyclometalation
[
1]
drogenation of ketones. Later, catalytic versions of this trans-
formation could be established with late-transition-metal com-
plexes containing ruthenium, iridium, or rhodium as the cata-
[8]
mechanism resulted in the in situ formation of a ruthenium-
coordinated carbanion. This was the key step to develop novel
moisture- and air-stable catalysts that allowed the transfer hy-
[
2]
lytically active sites. Noyori et al. developed the first asym-
metric transfer hydrogenations with high stereoselectivities.
They also discussed the potentials of a base-free hydrogena-
[9]
drogenation of ketones, and quite recently the reductive ami-
nation of aromatic aldehydes, to be performed under base-free
[
3]
tion process.
[10]
Despite these efforts, only a few base-free transfer hydroge-
nation reactions, catalyzed by iridium, iron, osmium, or rutheni-
conditions. However, this special feature has, to date, been
restricted to 2-(2-aminopyrimidin-4-yl)pyridine ligands with ter-
tiary amino groups attached to the pyrimidine ring. The posi-
tive effect of cyclometalation occurring at the (2-dialkylamino)-
pyrimidin-4-yl site on the performance of transition-metal cata-
lysts has also been observed for chelating ligands wherein
phosphines are combined with the (2-dialkylamino)pyrimidin-
[
4–7]
um complexes, have been described to date.
Some of them
require additives, such as silver salts, to activate the catalyst,
some are limited regarding the substrate scope, have long re-
action times, low reaction rates, or are highly sensitive towards
[
4c,5a,b,7d]
air and moisture.
Nevertheless, there are clear benefits
[11]
from working in the absence of a base: chiral compounds may
be prevented from undergoing racemization and corrosion can
be avoided.
4-yl moiety. We could synthesize highly active catalysts for
the palladium-catalyzed Suzuki–Miyaura coupling reaction by
applying such ligands.
Herein, we present a detailed mechanistic study on catalysts
6
+
of the type [(h -arene)Ru(X)(pympyr)] (pympyr=2-(pyrimidin-
4-yl)pyridine-type ligand) for the transfer hydrogenation of aryl
ketones with isopropanol as the hydrogen source (Scheme 1).
To gain a deeper insight into structure/reactivity relation-
ships, variations of the electronic and steric impact of a series
of differently substituted 2-pyrimidin-4-yl-pyridine ligands, as
[
a] Dr. C. Kerner, J. Lang, Dr. M. Gaffga, Dr. Y. Sun,
Prof. Dr. G. Niedner-Schatteburg, Prof. Dr. W. R. Thiel
Fachbereich Chemie, Technische Universitꢀt Kaiserslautern
Erwin-Schrçdinger-Strasse 54, 67663 Kaiserslautern (Germany)
E-mail: gns@chemie.uni-kl.de
thiel@chemie.uni-kl.de
[b] J. Lang, Dr. M. Gaffga, Prof. Dr. G. Niedner-Schatteburg
Landesforschungszentrum OPTIMAS Technische Universitꢀt Kaiserslautern
Erwin-Schrçdinger-Strasse 54, 67663 Kaiserslautern (Germany)
[c] Dr. F. S. Menges
Department of Chemistry,Yale University 225 Prospect St.,New Haven,
CT06511 (USA)
Supporting information for this article can be found under http://
dx.doi.org/10.1002/cplu.201600526.
Scheme 1. Transfer hydrogenation of aryl ketones with isopropanol as the
hydrogen source.
ChemPlusChem 2016, 81, 1 – 14
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
well as of the anionic coordinating ligand X and arene ligand,
were performed. The study is supported by mass spectrometry
and theoretical calculations.
Results and Discussion
Synthesis of ligands and ruthenium complexes
N,N’-Chelating 2-(pyrimidin-4-yl)pyridine ligands were synthe-
sized in a few steps starting with the condensation of 2-acetyl-
pyridine and N,N-dimethylformamide dimethyl acetal
(
Scheme 2), leading to 3-dimethylamino-1-pyridinylprop-2-en-
[
12]
1
-one (1), the central intermediate for all 2-(pyrimidin-4)yl-
pyridines (2) presented herein. The functionalized pyrimidine
moieties are subsequently generated by condensation of
2 2
Scheme 3. Synthesis of the ruthenium complexes 3a–iX: i) CH Cl , RT, 24 h.
1
with formamidinium acetate, which results in the formation
[
13]
of 2-(pyrimidin-4-yl)pyridine (2a). Alternatively, an appropri-
ate guanidinium salt leads to 2-(2-aminopyrimidin-4-yl)pyri-
[
9]
dines 2b–d. Condensation with urea gives 2-(2-hydroxypyri-
midin-4-yl)pyridine (2e), which can then be converted into the
corresponding chloro derivative 2 f.
the transfer hydrogenation. This effect is described in more
detail below. Apart from ligands 2a–f, commercially available
2g was coordinated to the ruthenium(II) center to give 3gX.
To study isotope effects occurring during the roll-over mecha-
nism, compound 2h was synthesized and coordinated to
ruthenium to form 3hX. To investigate the steric impact of the
6
h -coordinating arene ring on the transfer hydrogenation, one
example of a ruthenium complex with a benzene ligand (3iX)
was included.
Molecular structures
All ruthenium(II) complexes were characterized by means of
NMR spectroscopy and elemental analysis. Because the ruthe-
nium(II) site is a center of chirality in the case of complexes
3
a–fX, the resonances of the diastereotopic hydrogen and
6
carbon atoms of the h -coordinating cymene ring appear dou-
Scheme 2. Synthesis of the pyrimidinyl-functionalized chelating ligands:
i) 80 8C, 6 h, 67%; ii) formamidine acetate or guanidinium salt, Na, EtOH,
758C, 16 h, up to 66%; iii) urea, H
2
O/EtOH (18:3), 958C, concentrated HCl,
4
8 h, 85%; iv) POCl , 1058C, 1 h, 90%.
3
Although the cyclization of 1 with formamidinium and gua-
nidinium salts, which leads to ligands 2a–d, requires basic con-
[
9]
ditions, the condensation of 1 with urea, giving compound
e, has to be performed in the presence of an acid. The yield
2
for this transformation is high, as it is for the subsequent intro-
duction of a chloro substituent by treating 2e with POCl3.
To complete the series of ligands, 2,2’-bipyridine (2g) and
[
14]
[
D ]2,2’-bipyridine (2h) were also applied for coordination to
8
ruthenium(II), in addition to the N,N’-donors 2a–f described
6
above. By simply mixing [{(h -cymene)RuCl } ], one of the li-
2
2
ꢀ
gands 2a–h, and a salt of a weakly coordinating anion X in
dichloromethane, the desired ionic ruthenium(II) complexes of
6
the type [(h -cymene)(N,N’-ligand)RuCl]X 3a–hX are formed
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
(
Scheme 3). The counteranions X (BF4 , PF6 , BPh4 , BArF )
4
+
+
+
+
Figure 1. Molecular structures of the cations of 3a , 3d , 3 f , and 3i in
the solid state. The anions and cocrystallized solvent molecules are omitted
for clarity.
significantly influence the solubility of the cationic catalysts in
iPrOH, which is used as the solvent and hydrogen source for
&
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
bonds are directly influenced by the substitution motif at the
Table 1. Characteristic bond lengths [ꢁ], angles [8], and dihedral angles
[
a]
[b]
N,N’-coordinating ligand.
[
8] of 3aPF
6
,
3bBF
4
,
3dBF
4
, 3 fBF
4
, and 3iBF
4
.
3
aPF
6
3bPF
6
3dBF
4
3 fBF
4
3iBF
4
CID EIS-MS measurements
Ru1ꢀN1
Ru1ꢀN2
2.090(3)
2.080(3)
1.689(2)
2.398(1)
2.082(2) 2.081(2) 2.102(4)
2.120(2) 2.154(2) 2.152(3)
1.689(1) 1.690(1) 1.694(2)
2.395(1) 2.405(1) 2.391(1)
2.073(12)
2.159(12)
1.672(7)
2.418(4)
Further verification arose by means of collision-induced disso-
ciation (CID) EIS-MS measurements of isolated complexes. In
[
c]
Ru1ꢀAr
Ru1ꢀCl1
6
N1-Ru-N2
76.70(12) 76.55(7) 77.35(9) 76.73(14) 76.8(5)
2013, we reported a first but limited study on the two [(h -
+
+
N1-Ru1-Cl1
N2-Ru1-Cl1
N1-C5-C6-N2
86.91(9)
83.90(9)
0.6(6)
87.61(5) 82.13(7) 84.87(11) 83.5(3)
85.50(5) 89.57(6) 85.73(10) 92.6(3)
cymene)(N,N’-ligand)RuCl] cations 3a (tertiary amino group)
+
and 3d (primary amino group), as well as on an analogous
[
d]
0.7(3)
12.9(4)
5.6(6)
12.2(19)
system containing
a
tertiary (2-pyrrolidinylpyrimidinyl)
[a] 3aPF
6
also crystallized with the inclusion of one equivalent of KCl, see
[9]
moiety. It could clearly be shown that all three complexes un-
derwent fragmentation of HCl under CID EIS-MS conditions
and the proton originated from the ortho-CH moiety of the
pyrimidine ring. The latter fact was proved by selective deuter-
ation of this site.
the Supporting Information. [b] Data for comparison from Ref. [9]. [c] Dis-
tance between Ru1 and the centroid of the h -coordinated arene ligand.
6
[d] C5, C6: carbon atoms bridging the two heterocycles of the ligands.
However, the energies required for the fragmentation dif-
fered largely: it was much easier to detach HCl from the cat-
ions as long as a tertiary amino group was attached to the pyr-
imidine ring. Quantum chemical calculations performed on the
bled. Some of the new ruthenium complexes could be ob-
tained as single crystals suitable for an X-ray structure analysis.
Figure 1 shows the molecular structures of the cations of
+
+
3
3
aPF , 3dBF , 3 fBF , and 3iBF in the solid state. The data for
bPF6, which has already been published, is added to Table 1
three cations mentioned above and on 3a and 3g con-
6
4
4
4
[
9]
firmed these findings and demonstrated that the process fol-
[16]
for comparison with a compound containing a primary amino
group at the pyrimidine ring.
lowed a roll-over cyclometalation process (Scheme 4).
+
The whole series of cations 3a–i was investigated in the
gas phase to evaluate in detail the steric and electronic im-
pacts of different substituents on fragmentation. Figure 2
shows the recorded CID curves. The ELAB value at 50% reflects
the energy required to release HCl. In the following, it can be
expressed in terms of an activation energy by calibration with
For a derivative containing a secondary amino group
(
ArNH(nPr)) that is comparable to 3bBF , a Ru1ꢀN2 distance of
4
2
.108 ꢁ and a N1-C5-C6-N2 dihedral angle of 3.48 were
[
9]
found. The RuꢀN distances for compound 3gPF , which crys-
6
tallizes with two crystallographically independent molecules in
the unit cell, were reported to be in the range of 2.087–
[
15]
2
.101 ꢁ;
this is in agreement with the Ru1ꢀN1 distances
found in our study. The dihedral angle N1-C5-C6-N2 of 3gPF6
is close to 08. Together with previously published solid-state
[
9]
structures, the structural data of the new ruthenium(II) com-
plexes clearly prove the pronounced influence of the substitu-
ent attached in the 2-position of the 2-(pyrimidin-4-yl)pyridine-
type ligand on the Ru1ꢀN2 distance and, in particular, on the
dihedral angle N1-C5-C6-N2. While the distances between the
ruthenium(II) site Ru1 and the pyridine nitrogen atom N1 differ
only slightly in all ruthenium complexes, there is clearly
a broad variation in the Ru1ꢀN2 distances. Here steric and
electronic influences have to be considered. All heteroatomic
substituents weaken the RuꢀN2 bond by withdrawing electron
density (ꢀI effect) from the s framework of the ligand. This
electronic effect is superimposed by the steric interaction be-
tween these substituents and the p-coordinating aromatic
ligand, and thus, leads to a gradation of the N1-C5-C6-N2 dihe-
dral angles in the sequence 3aPF ꢁ3bPF <3 fBF <3iBF ꢁ
Scheme 4. Roll-over cyclometalation under CID ESI-MS conditions.
6
6
4
4
3
dBF . Even complex 3iBF , which contains a less sterically de-
4 4
6
manding h -benzene ligand, still adopts a dihedral angle of
2.218, which is similar to that in 3dBF . Here the distance be-
1
4
tween the metal and the centroid of the arene ligand is ap-
proximately 1.672 ꢁ, and thus, slightly shorter than those of
6
the h -cymene complexes (1.689–1.694 ꢁ). The Ru1ꢀCl1 bond
(
2.418 ꢁ) in 3iBF is a bit longer than in the four other com-
4
plexes (2.391–2.405 ꢁ). At this point, it can be summarized that
the bond parameters, and thus, strengths of the Ru1–ligand
Figure 2. Measured data (dots) and sigmoidal fit (lines) of the CID ESI-MS
curves for the cations 3a –3i .
+
+
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
+
Finally, there is cation 3i , which is the only compound con-
Table 2. Fitted ELAB values and relative reaction enthalpies for the split-
ting of HCl or DCl for cations 3a –3i .
6
+
+
taining a h -coordinating ligand other than the bulky cymene.
+
In comparison to 3d , a pronounced increase in the activation
Cation
E
LAB
DDHfit
[kJmol
DDHfit
[kcalmol ]
barrier was measured. This does, at a first glance, not attest to
the almost identical steric interference between the 2-(2-dime-
thylaminopyrimidin-4-yl)pyridine and arene ligands, as docu-
mented by the almost identical values of Ru1ꢀN2 and N1-C5-
ꢀ
1
ꢀ1
[V]
]
+
3
3
3
3
3
3
3
3
3
a
1.16
0.68
0.70
0.63
0.63
0.68
1.17
1.21
0.96
264
157
160
144
149
159
268
276
220
63
37
38
35
36
38
64
66
53
+
b
c
d
e
f
g
h
i
+
+
+
+
+
C6-N2 found for 3i and 3d . However, it has to be kept in
mind that solid-state structures deliver a static picture of a mol-
ecule, with generally minimized intramolecular steric interac-
tions. In solution and in the gas phase, the cymene ligand will
rotate on top of the ruthenium(II) site bringing its bulky iso-
+
+
+
+
propyl and methyl substituents closer to the NMe group of
2
the pyrimidine ring.
6
In our recent study, we showed that only [(h -arene)(2a)-
+
+
so-called thermometer ions (see the Supporting Informa-
RuCl] cations containing a tertiary amino group, such as 3b ,
[
17]
tion). Table 2 summarizes the resulting energies.
were able to catalyze the base-free transfer hydrogenation of
[9]
According to the CID ESI-MS data, the cationic ruthenium(II)
species can be divided into three groups: First, there are cat-
ketones. Even in the presence of a base (KOH), there are
large differences in the activities between compounds of the
+
+
+
+
+
ions 3a , 3g and 3h , which do not have any substituents
in the 2-position of each six-membered heterocyclic ring. They
show by far higher barriers for the roll-over cyclometalation
type 3d and 3b . Similar results were found for the reduc-
+
tive amination of benzaldehyde with anilines: compound 3d
catalyzes this reaction with high activity without requiring
+
+
[10]
than complexes 3b–3 f and 3i . Among the three high-barri-
er complexes, derivative 3a undergoes the roll-over cyclome-
talation somewhat easier than 3g and 3h . This can be ex-
plained by the presence of a second electronegative nitrogen
a base or any other additive.
+
+
+
Catalytic transfer hydrogenation
+
atom in the pyrimidine ring of 3a , which weakens the Ru1ꢀ With a whole series of essentially new ruthenium(II) com-
+
+
N2 bond. For species 3g and 3h , the expected primary iso-
pounds in our hands, we began a general survey on the struc-
ture/reactivity relationships in base-free transfer hydrogena-
tion. It should be noted at this point that a linear transfer of
the CID ESI-MS data (gas phase) into a quantitative reactivity
profile of the investigated class of compounds in solution is
not possible. However, the elucidation of trends concerning
the influence of the substitution pattern on the catalytic activi-
ty seems feasible. A series of parameters such as reaction tem-
perature, nature of the solvent and counteranion, and even
variations of the concentration may cause variations of the re-
activity.
[
18]
tope effect is observed. This speaks, according to Olah, for
the fact that the highest transition state is of a s-complex
nature, which is in agreement with the results of quantum
+
+
+
chemical calculations on compounds 3a , 3b , 3d , and
h , as recently published by us.
A second group of compounds (3b–3 f ) consists of ruthe-
nium(II) complexes with pyrimidinylpyridine ligands containing
a substituent with a heteroatom in the 2-position of the pyri-
midine ring. They undergo the roll-over cyclometalation much
easier than the group of compounds discussed above. Elec-
tronic and steric reasons may help to rationalize this observa-
tion: the electronegative heteroatom will withdraw electron
density from the s skeleton of the ligand. It furthermore will,
+
[9]
3
+
We therefore first investigated the influence of the counter-
+
anion on the reactivity of our most reactive catalyst 3d at
a quite high catalyst loading of 5 mol%. The conversion was
6
together with its substituents, sterically interfere with the h -
determined after a reaction time of 4 h at 828C. Table 3 sum-
cymene ligand. Both effects weaken the strength of the Ru1ꢀ marizes the results obtained with acetophenone (AC) as the
N2 bond, as documented by the solid-state structures dis-
cussed above, and thus, facilitate roll-over cyclometalation.
Within this group of compounds, further differences can be
determined: among the amino-functionalized derivatives, com-
model substrate. Entries 1–4 show the effects of the four coun-
teranions: complexes 3dBArF4 and 3dBF4 give the best re-
sults. However, there is no really dominant effect of the coun-
teranion on the catalytic activity, as that found for the nature
of the N,N’-donor ligand (see below). At these relatively high
catalyst concentrations, the observed variations in the activities
might be assigned to the solubility of the catalyst in iPrOH.
Attempts to form the catalyst in situ in the reaction mixture
+
+
pounds 3b (NH ) and 3c (NHMe) require higher activation
2
+
energies than 3d (NMe ). The variations in the bond lengths
2
Ru1ꢀN2 and torsion angles N1-C5-C6-N2 in the solid-state
+
+
structures of 3b and 3d are in accordance with this obser-
+
6
vation. The activation energy of hydroxyl derivative 3e is
of isopropanol and AC by simply adding [(h -cymene)RuCl ] ,
2
2
+
almost identical to the value of 3d . The stronger electron-
ligand 2d, and NaBPh in the correct molecular ratio lead to
4
withdrawing effect of the OH group seems to compensate for
no catalytic conversion at all (Table 3, entry 5). The strongly co-
ordinating solvent prevents the formation of the catalyst. From
this finding, the conclusion can be drawn that the preformed
ruthenium catalyst will not completely lose its N,N’-donor
the bulkiness of the NMe substituent. The electronically less
2
+
accepting chloro substituent in 3 f then increases the barrier
+
up to the value found for 3b .
&
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
activation barrier, gives the best results in catalytic transfer hy-
[
a]
Table 3. Screening of the base-free transfer hydrogenation of AC.
+
6
drogenation. However, derivative 3i , with a h -coordinated
benzene instead of a cymene ligand, still gives reasonable
yields, although the measured barrier is much higher than that
Entry
AC
mmol]
Cat.
[mol%]
Cat.
V(iPrOH)
[mL]
Yield [%]
(t [h])
[
+
1
2
3
4
0.5
0.5
0.5
0.5
0.5
0.5
1.0
1.0
1.0
1.0
1.0
1.0
1.0
1.0
0.5
1.5
1.4
5.0
5.0
5.0
5.0
5.0
5.0
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
0.5
3dBArF
4
25.0
25.0
25.0
25.0
25.0
25.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.0
5.3
88 (4)
71 (4)
44 (4)
for 3d ; this finding must relate to the CID process itself.
+
3dBPh
3dPF
3dBF
4
Chloro compound 3 f is still slightly active, although its acti-
6
+
+
vation barrier is comparable to those of inactive 3b and 3c ,
4
89 (4)
+
[
b]
c]
and hydroxyl derivative 3e is inactive, although its activation
5
6
7
8
9
1
3dBPh
3dBPh
3dBPh
4
4
4
0 (4)
[
+
56 (24)
97 (24)
53 (24)
0 (24)
10 (24)
0 (24)
barrier is as low as that found for 3d . The fact that deriva-
+
+
+
tives 3b , 3c , and 3e possessing polar NH or OH units are
catalytically inactive speaks for an additional influence of the
solvent. From the solid-state structures of compounds with pri-
3iBF
3eBF
3 fBF
3gPF
3aPF
3cPF
3bPF
4
4
0
4
1
1
6
mary (NH ) or secondary (NHR) amino groups on the pyrimi-
2
1
1
1
1
1
1
2
3
4
5
6
7
6
0 (24)
0 (24)
0 (24)
dine ring, it is known that there is no direct NꢀH···Cl hydrogen
bond. However, one molecule of isopropanol might bridge the
gap and provide an additional contribution to keep the ligand
in the N,N’-coordination mode (Scheme 5) by locking the N-C-
6
6
3dBF
3dBF
3dBF
4
4
4
76 (24)
90 (24)
100 (24)
[
a] T=828C, catalyst, AC, iPrOH, yields were determined by GC analysis
6
and are uncorrected. [b] 0.25 mol% of [(h -cymene)RuCl
ligand 2d, and 0.5 mol% NaBPh . [c] T=608C.
2 2
] , 0.5 mol% of
4
ligand during catalysis. Otherwise, no or only little product for-
mation should be observed.
According to CID ESI-MS measurements, the activation
+
energy for cation 3d is the lowest of all compounds investi-
gated. It therefore seemed to be feasible to lower the reaction
temperature and try to get some conversion. Indeed, this is
Scheme 5. Illustration of the interaction of one solvent molecule with the
amino moiety and chlorido ligand.
the case: even at 608C, compound 3dBPh provides a reasona-
4
ble yield of product (56%) after 24 h (Table 3, entry 6). To com-
pare the catalytic activities of ruthenium(II) complexes 3a–iX,
the catalyst loading was reduced to 0.5 mol%. Concurrently,
the solvent volume was decreased to 5 mL and the reaction
time was extended to 24 h (Table 3, entries 7–14). From the
whole series of 3a–iX, only compounds 3dBPh4 and 3iBF4
showed reasonable activities. Poor activity was observed for
C-N dihedral angle against further twisting. This would nicely
explain the catalytic inactivity of those compounds with protic
+
+
+
R groups (3b , 3c , 3e ).
To complete the investigations into transfer hydrogenations
with AC, the amount of substrate and the solvent volume
were optimized (Table 3, entries 15–17). The finally found opti-
mized conditions are described in Table 3, entry 17. They were
applied to the following elucidation of the substrate scope
3
fBF ; no activity at all for the other ruthenium(II) complexes.
4
There still might be a slight influence of the counteranion (see
discussion above), but the 2-(2-dimethylaminopyrimidin-4-yl)-
pyridine ligand with a tertiary amino group clearly boosts the
activity of the catalytically active metal site. It has to be men-
tioned that the compounds that are inactive under the base-
free conditions discussed herein become catalytically active as
soon as a base (KOH) is added to the reaction mixture. Howev-
er, they are still less active than derivatives with a tertiary
(Table 4) with 3dBF as the catalyst.
4
Generally, all substrates are converted into the correspond-
ing ketones under base-free conditions with good yields. There
are slight variations, mainly resulting from the steric impact of
the substitution pattern. The para-functionalized arylketones
with electron-withdrawing or -donating groups give high
yields after a reaction time of only 4 h, while the introduction
of a substituent in the ortho position results in lower conver-
sions. Alkylketones give even lower yields after 4 h. However,
regarding the performance after 24 h, there is not much differ-
ence between the substrates: yields of more than 80% are
found, which means that the catalyst stays intact for a long
time. For 2-hexanone, the yield of 91% after 24 h is close to
equilibrium.
[
9]
amino substituent.
Solid-state structural data is not available for all ruthenium
complexes, but for the seven derivatives discussed in the solid-
state structure section above there is a clear trend that relates
structural data with catalytic activities: A longer RuꢀN2 bond
and larger N1-C5-C6-N2 dihedral angle result in higher catalytic
activity. It seems that a dihedral angle of approximately 5–68,
as found for the chloro-substituted derivative 3 fBF , marks
4
Quantum chemical calculations
a border between active and inactive compounds. Regarding
the results of the CID ESI-MS study, a more differentiated view
From a mechanistic point of view, roll-over cyclometalation is
the only possible entrance into a base-free transfer hydrogena-
+
seems to be necessary. Cation 3d , which shows the lowest
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
+
on cation 3i . This allows us to avoid problems in locating
minima and transition states caused by the rotational flexibility
of the methyl, and especially isopropyl, group of the cymene
ligand. Figure 3 summarizes the calculations concerning roll-
Table 4. Substrate scope for the base-free transfer hydrogenation with
[
a]
3
dBF
4
as the catalyst.
[
a]
Entry
1
Substrate
Time [h]
4
Yield [%]
+
over cyclometalation occurring at cation 3i .
56
81
The calculated DDG values are close to the data obtained
2
2
2
4
+
[9]
for cation 3d . Additionally, critical structural parameters ob-
+
tained for the calculated molecular structure of 3i (denoted
4
4
84
96
2
3
4
in Figure 3 as A; for calculation details, see the Supporting In-
formation) are in very good agreement with respect to the
solid-state structure of 3iBF4 (calcd: Ru1ꢀN1 2.099, Ru1ꢀN2
2.177, Ru1ꢀCl1 2.402 ꢁ; N1-Ru1-N2 77.2, N1-Ru1-Cl1 81.3, N2-
Ru2-Cl1 90.9, N1-C5-C6-N2 16.48). Solely the distance Ru1ꢀAr
4
4
86
quant.
(
1.714 ꢁ) is slightly overestimated compared with the structure
4
4
89
quant.
of 3iBF . For the last transition state, the DDG value is slightly
4
2
2
lower than the DDG value of the preceding intermediate. This
is because optimization is performed on the DE hypersurface,
at which the last transition state is, as expected, higher in
energy than the preceding intermediate. On the DG hypersur-
face, an increase in entropy becomes noticeable, as caused by
the weekly bound HCl molecule.
4
4
84
98
5
4
4
82
quant.
6
7
2
2
Starting from the 16 VE intermediate D, the interaction with
isopropanol can occur in two different ways: either the OH
group attacks the ruthenium center or the hydrogen atom of
the central carbon atom of the isopropanol molecule coordi-
nates to ruthenium. This leads to two different reaction se-
quences that both finally give the same hydrido ruthenium
species.
4
4
74
95
4
4
74
95
8
9
2
2
4
4
56
75
Attack of D by the isopropanol hydrogen atom (Figure 4)
can be considered as part of an outer-sphere mechanism,
wherein the isopropanol molecule does not coordinate to the
ruthenium(II) site through the oxygen atom. The resulting in-
4
4
47
84
1
0
2
2
ꢀ
1
termediate E is endergonic by approximately 10.6 kcalmol ,
4
4
55
91
ꢀ1
11
but exothermic by approximately 1.3 kcalmol with respect to
16 VE intermediate D and free isopropanol. This indicates that
[a] 1.42 mmol of ketone, 5.3 mL (69.2 mmol) of iPrOH, 0.5 mol% of
the stabilization of the 16 VE system by the Ru···H contact is
only weak.
3
dBF , 828C. [b] Yields determined by GC analysis and are corrected.
4
The transition state connecting the 16 VE system D with this
first intermediate E could not be located. Subsequently, the
formation of the RuꢀH unit is accompanied by the transfer of
a proton from the isopropanol OH group to the most basic
site (carbanion) of the ligand, which allows the energetically
unfavorable generation of charged species to be avoided.
However, transition state EF of the concerted hydride/proton
transfer is quite high in energy, which makes this route unfav-
orable. The reason for this high barrier can be elucidated from
the structure of EF: the formation of the RuꢀH bond is nearly
accomplished, but the OꢀH bond is still far from being broken.
This means that the structure of the substrate resembles a pro-
tonated molecule of acetone with a high accumulation of posi-
tive charge on the carbonyl carbon atom. Roll-over cyclodeme-
talation then takes place via transition state FG, leading to cat-
ionic ruthenium(II) hydrido complex G with a strongly bent
backbone of the N,N’-donor, similar to the structure found for
tion with this type of ruthenium(II) complex. According to CID
[
9]
ESI-MS data (see above) and DFT calculations, it is especially
favorable for 2-(pyrimidin-4-yl)pyridine ligands with a tertiary
amino group in the 2-position of the pyrimidine ring because
it weakens the Ru1ꢀN2 bond by steric and electronic factors.
The breaking of this bond is the first step in the roll-over cyclo-
metalation process, which finally leads to a 16 valence electron
(VE) cation in the gas phase. In solution, it is assumed that iso-
propanol will attack at the vacant coordination site and will
subsequently undergo the transfer of a hydride anion to form
a ruthenium hydrido species; the central intermediate in the
transfer hydrogenation cycle. To date, there have been no the-
oretical calculations available for a more detailed understand-
ing of this second part, the formation of the hydrido inter-
mediate. This is discussed in the following section.
6
+
Because compound 3iBF , with a h -coordinated benzene
3i (A). This ruthenium(II) hydrido complex G is only slightly
4
ꢀ
1
instead of a cymene ligand, shows similar catalytic activities to
endergonic (3.9 kcalmol ), but rather endothermic (16.7 kcal
those of 3dBF , it seemed feasible to perform DFT calculations
4
&
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
+
Figure 3. Calculated pathway for the roll-over cyclometalation of 3i (DDG given).
Figure 4. Calculated pathway for a concerted hydride/proton transfer leading to a ruthenium(II) hydrido species (DDG given).
ꢀ
1
+
mol ) with respect to 3i (A), which explains why it cannot
be observed in isopropanol.
ed. It is quite remarkable that there is only a slight stabilization
ꢀ
1
of approximately 2.5 kcalmol between final product J and
the preceding s-complex I. Because the central barrier HI
By another conceivable route to the ruthenium(II) hydrido
complex (G), the substrate isopropanol would attack the ruthe-
nium(II) site by the oxygen atom. This sequence can be divided
into two sections: first, the formation of the ruthenium(II) iso-
propanolato complex J (Figure 5); second, the formation of the
hydrido complex G (Figure 6). The formation of the isopropa-
nolato complex follows a roll-over cyclodemetalation mecha-
nism, and thus, resembles in its intermediates and transition
ꢀ
1
(38.7 kcalmol ), which corresponds to the protonation of the
carbanion coordinated to ruthenium, is approximately 6.7 kcal
ꢀ
1
mol higher than that for the HCl splitting process (32.0 kcal
ꢀ
1
mol ), isopropanolato complex J should not be detectable in
the reaction mixture, as long as the barrier for subsequent hy-
dride transfer is low enough.
This hydride transfer (Figure 6) is in accordance with a b-H
elimination from the isopropanolato ligand, and therefore, re-
quires partial de-coordination of the N,N’-donor. Otherwise, b-
+
states the splitting of HCl from 3i (A) in the reverse direction.
The first (DH) and last (IJ) transition states could not be locat-
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Figure 5. Calculated pathway for the formation of a ruthenium(II) isopropanolato complex (DDG given).
Figure 6. Calculated pathway for the formation of a ruthenium(II) hydrido from an isopropanolato complex (DDG given).
H elimination would proceed via an energetically disfavored 20
VE species. Alternatively, the arene ligand could undergo
lower in energy than that of KF. The overall barrier for this pro-
ꢀ
1
cess is at approximately 35.0 kcalmol . It therefore is indeed
lower than the barrier for the formation of the ruthenium(II)
isopropanolato complex discussed above.
6
4
a h !h shift, but the calculations do not give any indication
of such a process.
Starting from the second to last structure, I, shown in
Figure 5, an agostic b-H interaction in intermediate K is ena-
bled through the de-coordination of the weakly s-bound
carbon atom of the pyrimidine ring. A seemingly isoenergetic
transition state (IK) could not be located. We assume that this
barrier is slightly above the b-H bound intermediate K. Only
a tiny barrier of 0.4 kcalmol (KF) is required to release ace-
tone from this structure and to form ruthenium(II) hydrido spe-
cies G via intermediate F and transition state FG, which is
From the results of the calculations, a series of conclusion
can be drawn. The crucial hydrido (G) and isopropanolato (J)
intermediates are higher in energy than the chlorido derivative
+
3i (A), which is in agreement with the fact that these inter-
mediates cannot be observed in isopropanol. The formation of
hydrido intermediate G most likely does not follow an outer-
sphere mechanism, but instead a classical b-H elimination
takes place. The barrier (KF) for b-H elimination, which requires
partial de-coordination of the N,N’-ligand, is at approximately
ꢀ
1
&
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
ꢀ
1
3
5.0 kcalmol , and therefore, much lower than the barrier
Synthesis
found for the outer-sphere hydrogen transfer (EF). However, it
is in the same range as that of barrier BC, which is required for
roll-over cyclometalation starting from 3i . This is not surpris-
Compound 2c: Methylguanidinium chloride (1.97 g, 18.0 mmol)
and sodium (3.0 g) were dissolved in anhydrous ethanol (34 mL).
Compound 1 (3.17 g, 18.0 mmol) was added to the resulting sus-
pension, which was then heated to reflux for 16 h. After cooling to
RT, the solvent was removed under reduced pressure and the resi-
due was washed with cold water. The product was dried under
vacuum and the resulting brown oil was purified by column chro-
+
ing because the reverse b-H elimination is simply a roll-over
cyclometalation starting from the N,N’-coordinated hydrido
complex G. This fact strongly supports the idea that the activa-
+
tion of 3i proceeds through a roll-over cyclometalation pro-
cess. The hydrido intermediate G is the starting point for the
transfer of hydrogen to the ketone, which returns to com-
pound H (G!F!K!I!H) as shown in Figures 6 and 5.
matography (CH
Cl /MeOH) to give the product as a pale yellow
2 2
solid (13%). M.p. 108.08C; IR (ATR,): ~n =3359 (w), 3062 (w), 2900
ꢀ1
1
(w), 1579 (s), 1545 (s), 1393 (s), 778 cm (vs); H NMR (400.1 MHz,
3
CDCl , 208C): d=8.65 (dd, J(H,H)=0.68, 4.71 Hz, 1H; H1), 8.41 (d,
3
3
3
J(H,H)=5.11 Hz, 1H; H8), 8.35 (d, J(H,H)=7.96 Hz, 1H; H4), 7.76
3
3
Conclusion
(td, J(H,H)=1.75, 7.76 Hz, 1H; H3), 7.53 (d, J(H,H)=5.10 Hz, 1H;
3
H7), 7.31 (ddd, J(H,H)=1.09, 4.79, 7.49 Hz, 1H; H2), 5.61 (brs, 1H;
3
13
1
Weakening a critical ligand-to-metal bond by a combination of
electronic and steric factors results in roll-over cyclometalation
H10), 3.02 ppm (d, J(H,H)=5.04 Hz, 3H; H11);
(100.6 MHz, CDCl , 208C): d=163.7 (C6), 163.3 (C9), 159.3 (C8),
154.9 (C5), 149.5 (C1), 137.0 (C3), 125.1 (C2), 121.5 (C4), 106.9 (C7),
C{ H} NMR
3
6
at some of the ionic ruthenium(II) complexes of the type [(h -
2
8.6 ppm (C11); elemental analysis calcd (%) for C H N (186.22):
10 10 4
arene)(N,N’-ligand)RuCl]X (N,N’=N,N’-chelating 2-(pyrimidin-4-
yl)pyridine ligand). This crucial step allows the base-free trans-
fer hydrogenation of ketones under mild and base-free condi-
tions, as long as a tertiary amino group is attached to the 2-
position of the pyrimidinyl ring. Because we could determine
this effect for combinations of pyrimidines with two other
donor sites (pyridines and phosphanes), it more and more be-
comes clear that it is of general importance for this type of
donor site. It furthermore becomes clear that roll-over cyclo-
metalation could provide an electronic situation at a catalytical-
ly active site that is as beneficial for catalytic reactions as it is
for the photophysical properties of certain late-transition-metal
complexes. We are presently working out protocols for other
catalytic reactions under inclusion of the roll-over cyclometala-
tion process.
C 64.50, N 30.09, H 5.41; found: C 64.54, N 29.99, H 5.44.
Compound 2e: Urea (5.11 g, 85.1 mmol) and 1 (6.00 g, 34.0 mmol)
were dissolved in a mixture of H O/EtOH (18:3 mL). After the mix-
ture was heated to 958C, concentrated HCl (18.2 mL) was added.
The reaction mixture was kept at 958C for 48 h. After cooling to
RT, the solution was neutralized by the addition of a 2m solution
of Na CO . The precipitate was filtered off, washed with Et O, and
dried under vacuum. The product was isolated as a gray solid
(85%). M.p. 250.78C; IR (ATR): ~n =3017 (w), 2696 (m), 1638 (vs),
1611 (vs), 1432 (s), 1423 (s), 781 cm (s); H NMR (400.1 MHz,
2
2
3
2
ꢀ1
1
3
[D ]DMSO, 208C): d=12.03 (brs, 1H; H10), 8.74 (ddd, J(H,H)=0.83,
6
3
1
.62, 4.70 Hz, 1H; H1), 8.32 (d, J(H,H)=6.42 Hz, 1H; H8), 8.10 (d,
3
3
J(H,H)=6.42 Hz, 1H; H4), 8.01 (td, J(H,H)=1.76, 7.75 Hz, 1H; H3),
3
7
.59 (ddd, J(H,H)=1.16, 4.75, 7.54 Hz, 1H; H7), 7.31 ppm (d,
3
13
1
J(H,H)=6.43 Hz, 1H; H2); C{ H} NMR (100.6 MHz, [D ]DMSO,
6
2
1
08C): d=159.7 (C6), 156.8 (C9), 152.8 (C8), 149.6 (C5), 148.4 (C1),
37.6 (C3), 126.4 (C2), 121.8 (C4), 100.0 ppm (C7); elemental analy-
Experimental Section
sis calcd (%) for C H N O (173.17): C 62.42, N 24.26, H 4.07; found:
9
7
3
General
C 62.43, N 23.81, H 4.11.
Organic precursors were purchased form Sigma–Aldrich, Acros,
and TCI Chemicals and used without further purification. Com-
Compound 2 f: Compound 2 (4.00 g, 23.1 mmol) were added to
[12]
[13]
[9a]
[9a]
[14]
[19]
6
POCl
POCl
3
(40 mL). The mixture was heated to reflux for 1 h. Unreacted
was removed trap-by-trap and the residue was poured onto
pounds 1, 2a, 2b, 2d, 2h, 3gPF6, and bis[(h -benze-
[20]
3
ne)ruthenium(II)dichloride] were synthesized according to pub-
6
ice (10.0 g). After neutralization by the addition of a saturated solu-
tion of Na CO , the precipitated solid was filtered off and dried
lished procedures. The ruthenium(II) precursor [{(h -cymene)RuCl } ]
2
2
2
3
was commercially available from Strem Chemicals. The solvents
CH Cl and Et O were purchased from Sigma Aldrich and dried
under vacuum. The product was isolated as a gray solid (90%).
M.p. 107.78C; IR (ATR): ~n =3067 (w), 1565 (m), 1535 (s), 1423 (m),
2
2
2
over molecular sieves in a Braun MB SPS solvent dryer. Alcohols
were dried by distillation over magnesium. Isopropanol was pur-
chased from Bernd Kraft GmbH. The starting materials for catalysis
were purified by standard purification methods. All reactions were
performed under an atmosphere of nitrogen by using standard
Schlenk techniques. The NMR spectra were obtained on Bruker
Avance 400 and 600 systems with CDCl , CD CN, or [D ]DMSO as
ꢀ
1
1
1
2
0
(
4
345 (s), 1181 (s), 763 cm (vs); H NMR (400.1 MHz, [D ]DMSO,
6
3
3
08C): d=8.92 (d, J(H,H)=5.13 Hz, 1H; H1), 8.79 (dd, J(H,H)=
.75, 4.70 Hz, 1H; H8), 8.38 (t, J(H,H)=6.90 Hz, 2H; H3, H4), 8.05
3
3
3
td, J(H,H)=1.75, 7.77 Hz, 1H; H7), 7.64 ppm (ddd, J(H,H)=1.10,
13
1
.74, 7.56 Hz, 1H; H2); C{ H} NMR (100.6 MHz, [D ]DMSO, 208C):
6
d=165.4 (C6), 161.9 (C9), 160.4 (C8), 151.6 (C5), 150.0 (C1), 138.0
C3), 126.8 (C2), 121.8 (C5), 116.2 ppm (C7). elemental analysis
calcd (%) for C H ClN (191.62): C 56.41, H 3.16, N 21.93; found: C
3
3
6
(
the solvents. For the assignment of the signals, see the structures
presented in the Supporting Information. Preparative chromato-
graphic purifications were performed with a Teledyne Isco Combi
Flash Rf200 system by using prepacked silica columns from Redi-
Sep. The IR spectra were recorded with a PerkinElmer FTIR Spec-
trum 100 device equipped with an ATR sample assembly. For data
refining, the PerkinElmer software Spectrum 6.3.5 was used. Ele-
mental analysis measurements were performed with a Hanau Ele-
mental Analyzer Vario Microcube.
9
6
3
5
6.49, H 3.22, N 21.58.
General procedure for the synthesis of the ruthenium com-
[9]
6
plexes : [(h -Cymene)RuCl ] , the N,N’-chelating ligand, and a salt
2
2
providing the desired anion were added to a Schlenk tube. CH Cl
2
2
was added and the resulting suspension was stirred for 24 h at RT.
Then the reaction mixture was filtered, the amount of solvent was
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
reduced, and the desired complex was crystallized by the diffusion
(%) for C H BClF N Ru (543.76): C 44.18, H 4.45, N 10.30; found: C
20
24
4
4
of Et O.
43.82, H 4.58, N 10.03.
2
Compound 3aPF : The reaction of 2a (138 mg, 0.88 mmol), KPF
Compound 3dBArF : The reaction of 3d (64.1 mg, 0.32 mmol),
6
6
4
6
6
(
177 mg,
0.96 mmol),
and
[{(h -cymene)RuCl } ]
(245 mg,
NaBArF (399 mg, 0.45 mmol), and [{(h -p-cymene)RuCl } ] (91.9 mg,
2
2
4
2 2
0
.40 mmol) in CH Cl (70 mL) gave the product as an orange
0.15 mmol) in CH Cl2 (25 mL) gave the product as red crystals
2
2
2
powder (30%). M.p. 1888C; IR (ATP): ~n =3069 (w), 2974 (w), 1597
(93%). M.p. 1308C; IR (ATR): ~n =2962 (w), 1596 (m), 1555 (m), 1353
(s), 1272 (vs), 1111 (vs), 886 cm (s); H NMR (400.1 MHz, CDCl ,
3
ꢀ
1
1
ꢀ1
1
(
m), 1466 (m), 1155 (m), 827 (vs), 798 cm (s); H NMR (600.1 MHz,
3
3
3
CD CN, 208C): d=9.92 (s, 1H; H9), 9.40 (d, J(H,H)=5.41 Hz, 1H;
H1), 9.10 (d, J(H,H)=5.27 Hz, 1H; H8), 8.46 (d, J(H,H)=7.98 Hz,
208C): d=8.93 (d, J(H,H)=5.23 Hz, 1H; H1), 8.46 (d, J(H,H)=
3 4
3
3
3
4.56 Hz, 1H; H8), 7.90 (td, J(H,H)=7.88 Hz, J(H,H)=1.37 Hz, 1H;
H4), 7.82 (d, J(H,H)=7.59 Hz, 1H; H3), 7.70 (brs, 8H; HBArF), 7.55
(ddd, J(H,H)=5.68, 7.32 Hz, J(H,H)=1.36, 1H; H2), 7.51 (brs, 4H;
), 7.13 (d, J(H,H)=4.60 Hz, 1H; H7), 5.47 5.35 5.31 5.28 (4ꢂd,
4ꢂ J(H,H)=6.14, 6.16, 6.40, 6.27 Hz, 1H; Hcym), 3.41 (brs, 6H; H10),
2.43 (hept, J(H,H)=6.93 Hz, 1H; CH(CH ) ), 2.08 (s, 3H; CH ), 1.04,
0.89 ppm (2ꢂd, 2ꢂ J(H,H)=6.90, 6.96 Hz, 3H; CH(CH3)2);
3
3
1
H; H4), 8.27 (dd, J(H,H)=0.92, 5.28 Hz, 1H; H7), 8.25 (td,
3
3
4
J(H,H)=1.32, 7.91 Hz, 1H; H3), 7.82 (m, 1H; H2), 6.00 (2ꢂd, 2ꢂ
3
3
3
J(H,H)=6.58, 6.58 Hz, 2ꢂ1H; Hcym), 5.82 (2ꢂd, 2ꢂ J(H,H)=6.79,
H
BArF
3
6
1
.79 Hz, 2ꢂ1H; Hcym), 2.69 (sept, 1H; CH(CH ) ), 2.20 (s, 3H; CH ),
3 2
3
3
3
.06, 1.04 ppm (2ꢂd, 2ꢂ J(H,H)=7.14, 6.96 Hz, 2ꢂ3H; CH(CH ) );
3
2
3 2
3
1
3
1
3
C{ H} NMR (150.9 MHz, CD CN, 208C) d=163.4 (C9), 162.2 (C6),
3
13
1
1
1
61.1 (C8), 157.0 (C1), 153.9 (C5), 141.2 (C3), 130.6 (C2), 126.7 (C4),
C{ H} NMR (100.6 MHz, [D ]DMSO, 208C) d=168.3 (C9), 163.7 (C8),
6
20.0 (C7), 107.3, 104.6, 87.3, 86.7, 85.7, 85.4 (6ꢂC ), 31.9
161.1 (m, Cmeta), 158.5 (C6), 156.1 (C1), 154.0 (C5), 140.1 (C3), 134.1
cym
19
1
(CH(CH ) ), 22.2, 22.2 (2ꢂCH(CH3)2), 18.9 ppm (CH3); F NMR
(s, Cipso), 128.5 (m, Cortho), 128.4 (C2), 125.4 (C4), 124.0 (q, J =
3
2
FC
1
(
376.5 MHz, CD CN, 208C): d=ꢀ72.9 ppm (d, J(P,F)=706 Hz);
272.4 Hz, CF ), 117.4 (m, Cpara), 109.9 (C7), 103.7, 102.1, 85.4, 85.2,
3
3
3
1
1
P{ H} NMR (150.9 MHz, CD CN, 208C): d=12.9 ppm (hept); ele-
84.2, 83.3 (6ꢂCCym), 31.7 (C10), 25.0 (CH(CH ) ), 21.9 21.0 (2ꢂ
3
3 2
19
mental analysis calcd (%) for C H ClF N PRu (574.9): C 39.83, H
CH(CH ) ), 17.7 ppm (CH ); F NMR (376.5 MHz, CD CN, 208C): d=
19
23
6
3
3 2
3
3
11
3
.69, N 7.33; found: C 39.71, H 3.72, N 7.29.
ꢀ62.3 ppm; B NMR (128.4 MHz, CD CN, 208C): d=ꢀ6.63 ppm; el-
3
emental analysis calcd (%) for C H BClF N Ru (1334.21): C 47.71,
53
38
24
4
H 2.87, N 4.20; found: C 47.74, H 3.17, N 4.00.
Compound 3bPF : The reaction of 2b (74.3 mg, 0.43 mmol), KPF
6
6
6
(
95.3 mg, 0.52 mmol), and [{(h -cymene)RuCl } ] (132 mg,
2
2
0
.22 mmol) in CH Cl (10 mL) gave the product as an orange
Compound 3eBF : The reaction of 3e (218 mg, 1.26 mmol), NaBF
2
2
4
4
6
powder (63%). M.p. 2228C; IR (ATP): ~n =3438 (w), 3294 (w), 3095
(145 mg, 1.32 mmol), and [{(h -p-cymene)RuCl } ] (367 mg,
2
2
ꢀ
1
1
(
w), 1713 (w), 1557 (m), 823 (vs), 772 cm (s); H NMR (400.1 MHz,
0.6 mmol) in CH Cl2 (30 mL) gave the product as an orange
2
3
CD CN, 208C): d=9.28 (d, J(H,H)=5.61 Hz, 1H; H1), 8.61 (d,
powder (87%) from CH CN and Et O. M.p. 1978C; IR (ATR): ~n =3061
3
3
2
3
3
ꢀ1
J(H,H)=4.82 Hz, 1H; H8), 8.28 (d, J(H,H)=7.69 Hz, 1H; H4), 8.18
(m), 1671 (vs), 1612 (vs), 1600 (vs), 1225 (vs), 840 (vs), 776 cm
3
3
1
(
t, J(H,H)=7.67 Hz, 1H; H3), 7.75 (t, J(H,H)=6.12 Hz, 1H; H2), 7.53
(vs); H NMR (400.1 MHz, CD CN, 208C): d=10.92 (brs, 1H; H10),
3
3 3
3
(
d, J(H,H)=4.89 Hz, 1H; H7), 6.51 (brs, 2H; H10), 6.05, 5.87 (2ꢂd,
9.43 (d, J(H,H)=5.42 Hz, 1H; H1), 8.36 (d, J(H,H)=8.13 Hz, 1H;
3
3
3
2
1
6
2
1
8
1
ꢂ J(H,H)=6.04, 6.08 Hz, 1H; Hcym), 5.67 (m, 2H; Hcym), 2.52 (sept,
H4), 8.21 (t, J(H,H)=7.81 Hz, 1H; H3), 8.13 (d, J(H,H)=6.56 Hz,
1H; H8), 7.81 (m, 1H; H2), 7.18 (d, J(H,H)=6.72 Hz, 1H; H7), 6.14,
5.98 (2ꢂd d, 2ꢂ J(H,H)=5.98, 6.01 Hz, 1H; H ), 5.79 (m, J(H,H)=
6.27 Hz, 2H; Hcym), 2.64 (hept, J(H,H)=6.76 Hz, 1H; CH(CH ) ), 2.20
(s, 3H; CH ), 1.01 ppm (2ꢂd, J(H,H)=6.76 Hz, 6H; CH(CH3)2);
C{ H} NMR (100.6 MHz, CD CN, 208C) d=171.6 (C9), 157.2 (C1),
3
3
H; CH(CH ) ), 2.23 (s, 3H; CH ), 1.01, 0.93 ppm (2ꢂd, 2ꢂ J(H,H)=
3
2
3
13
1
3
3
.89, 6.85 Hz, 2ꢂ3H; CH(CH ) ); C{ H} NMR (100.6 MHz, CD CN,
3
2
3
cym
3
08C) d=165.9 (C9), 162.5 (C8), 162.3 (C6), 156.8 (C1), 155.0 (C5),
41.1 (C3), 130.0 (C2), 126.1 (C4), 110.0 (C7), 106.6, 106.2, 87.8, 86.6,
4.8, 84.5 (6ꢂC ), 31.7 (CH(CH ) ), 22.2, 22.0 (2ꢂCH(CH ) ),
3 2
3
3
13
1
cym
9
3 2
3 2
3
1
8.9 ppm (CH ); F NMR (376.5 MHz, CD CN, 208C): d=ꢀ73.0 ppm
154.7 (C6), 154.1 (C5), 150.7 (C3), 140.9 (C4), 130.9 (C2), 128.7 (C8),
3
3
1
31
1
(
d, J(P,F)=706 Hz); P{ H} NMR (162.0 MHz, CD CN, 208C): d=
106.5, 105.6, 101.9, 88.5, 86.5, 85.3, 84.1 (6ꢂCcym), 31.9 (CH(CH3)2),
3
19
ꢀ
146.7 ppm (sept); elemental analysis calcd (%) for
22.4, 22.3 (2ꢂCH(CH ) ), 19.3 ppm (CH ); F NMR (376.5 MHz,
3 2 3
1
1
C H ClF N PRu (587.9): C 38.82, H 3.77, N 9.53; found: C 38.77, H
CD CN, 208C): d=ꢀ151.8 ppm; B NMR (138.4 MHz, CD CN, 208C):
1
9
22
6
4
3
3
3
.93, N 9.23.
d=ꢀ1.18 ppm; elemental analysis calcd (%) for C H BClF N ORu
19
21
4
3
(
530.72): C 43.00, H 3.99, N 7.92; found: C 42.76, H 4.09, N 7.87.
Compound 3 fBF : The reaction of 2 f (241 mg, 1.26 mmol), NaBF
4
Compound 3cBF : The reaction of 2c (70.0 mg, 0.38 mmol), NaBF
4
4
6
(
69.0 mg, 0.63 mmol), and [{(h -cymene)RuCl } ] (96.3 mg,
2
2
4
6
0
.16 mmol) in CH Cl (30 mL) gave the product as an orange
(145 mg, 1.32 mmol), and [{(h -p-cymene)RuCl } ] (367 mg,
2
2
2 2
powder (69%). M.p. 1998C (decomp); IR (ATP): ~n =3394 (w), 3095
0.60 mmol) in CH Cl (30 mL) gave the product as a brown–orange
2
2
ꢀ1
1
(
(
8
w), 2974 (w), 1596 (m), 1574 (d), 1057 (vs), 1033 cm (vs); H NMR
solid (45%). M.p. 2008C; IR (ATR): ~n =3049 (w), 2970 (w), 1587 (m),
3
ꢀ1
1
600.1 MHz, CD CN, 208C): d=9.46 (d, J(H,H)=5.29 Hz, 1H; H1),
1050 (vs), 1039 (vs), 795 (s), 760 cm (s); H NMR (400.1 MHz,
3
3
3
3
4
.64 (d, J(H,H)=4.77 Hz, 1H; H8), 8.41 (d, J(H,H)=7.96 Hz, 1H;
CD CN, 208C): d=9.33 (dd, J(H,H)=5.56 Hz, J(H,H)=0.65 Hz, 1H;
3
3 3
3
H4), 8.16 (d, J(H,H)=7.70 Hz, 1H; H3), 7.75 (m, 1H; H7), 7.60 (d,
H1), 8.95 (d, J(H,H)=5.16 Hz, 1H; H8), 8.45 (d, J(H,H)=8.07 Hz,
3
3
J(H,H)=4.83 Hz, 1H; H2), 6.50 (d, J(H,H)=4.55 Hz, 1H; H10), 6.06,
1H; H4), 8.26 (m, 2H; H3, H7), 7.85 (m, 1H; H2), 6.16, 5.91, 5.81,
3
3
5
.94, 5.89, 5.87 (4ꢂd, 4ꢂ J(H,H)=6.05, 6.02, 6.02, 6.08 Hz, 4ꢂ1H;
5.76 (4ꢂd, 4ꢂ J(H,H)=6.14, 5.47, 6.25, 6.16 Hz, 1H; Hcym), 2.61
3
3
Hcym), 3.17 (d, J(H,H)=4.85 Hz, 3H; H11), 2.48 (sept, 1H; CH(CH ) ),
(hept, J(H,H)=6.96 Hz, 1H; CH(CH ) ), 2.26 (s, 3H; CH ), 1.01,
3 2
3 2
3
3
3
2
2
1
1
3
.22 (s, 3H; CH ), 0.97, 0.89 ppm (2ꢂd, 2ꢂ J(H,H)=6.90, 6.94 Hz,
0.94 ppm (2ꢂd, 2ꢂ J(H,H)=6.93, 6.93 Hz, 3H; CH(CH3)2);
C{ H} NMR (100.6 MHz, CD CN, 208C): d=165.5 (C9), 165.3 (C6),
3
1
3
13
1
ꢂ3H; CH(CH ) ); C NMR (50.3 MHz, CD CN, 208C) d=164.5 (C9),
3
2
3
3
62.6 (C8), 161.9 (C6), 157.3 (C1), 155.1 (C5), 141.0 (C3), 130.0 (C2),
26.3 (C4), 109.0 (C7), 106.2, 106.0, 87.8, 86.9, 85.0, 84.6 (6ꢂCcym),
1.7 (C11), 30.1 (CH(CH ) ), 22.4, 22.2 (2ꢂCH(CH ) ), 18.9 ppm (CH );
162.6 (C8), 157.1 (C1), 153.5 (C5), 141.4 (C3), 131.2 (C2), 127.6 (C4),
118.5 (C7), 108.0, 107.5, 88.9, 87.0, 86.1, 83.0 (6ꢂC ), 31.9
cym
1
9
(CH(CH3)2), 22.5, 22.2 (2ꢂCH(CH3)2), 19.2 ppm (CH3); F NMR
3
2
3 2
3
1
9
11
11
F NMR (376.5 MHz, CD CN, 208C): d=ꢀ151.8 ppm; B NMR
(376.5 MHz, CD CN, 208C): d=ꢀ151.8 ppm; B NMR (138.4 MHz,
3
3
(
128.4 MHz, CD CN, 208C): d=ꢀ1.18 ppm; elemental analysis calcd
CD CN, 208C): d=ꢀ1.19 ppm; elemental analysis calcd (%) for
3
3
&
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
10
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
C H BCl F N Ru (549.17): N 7.65, C 41.56, H 3.67,; found: N 7.54, C
ESI-MS experiments
1
9
20
2
4
3
4
1.35, H 3.74.
ESI-MS and CID measurements were performed by using a Paul-
type quadrupole ion trap instrument (Bruker Esquire 3000plus).
The ion source was set to positive and negative ESI modes. The
Compound 3hPF : The reaction of 2h (74.2 mg, 0.45 mmol), KPF
6
6
6
(
82.8 mg, 0.45 mmol), and [{(h -p-cymene)RuCl } ] (122 mg,
2
2
ꢀ1
scan speed was 13000 m/zs in standard resolution scan mode
0
.20 mmol) in CH Cl (20 mL) gave the product as a yellow powder
2 2
(0.3 full-width at half-maximum (FWHM)/m/z) and the scan range
(
(
(
13%). M.p. 2198C (decomp); IR (ATR): ~n =2971 (w), 2933 (w), 1581
w), 1338 (m), 1236 (m), 878 (m), 827 cm
400.1 MHz, CD CN, 208C): d=5.91, 5.71 (2ꢂd, 2ꢂ J(H,H)=6.43,
ꢀ
1
1
was m/z 15–1200. Mass spectra were accumulated for at least
(vs); H NMR
n
3
2 min. MS spectra were accumulated for at least 20 s. Sample solu-
3
3
tions were continuously infused into the ESI chamber by a syringe
6
.42 Hz, 2H; Hcym), 2.64 (hept, J(H,H)=6.92 Hz, 1H; CH(CH ) ), 2.19
3 2
ꢀ1
3
pump at a flow rate of 2 mLmin . Nitrogen was used as the drying
(
s, 3H; CH3), 1.02 ppm (d, J(H,H)=6.94 Hz, 6H; CH(CH3)2);
ꢀ1
1
3
1
gas at a flow rate of 3.0 Lmin at 2208C. The solutions were
C{ H} NMR (100.6 MHz, CD CN, 208C) d=156.1 (m, C1), 155.6 (C5),
3
sprayed at a nebulizer pressure of 280 mbar (4 psi) and the electro-
spray needle was held at 4.5 kV.
1
40.4 (m, C4), 128.1 (m, C3), 124.4 (m, C2), 106.0, 104.5, 87.4, 85.4
19
(
Ccym), 31.8 (CH(CH ) ), 22.1 (CH(CH ) ), 18.9 ppm (CH ); F NMR
3 2 3 2 3
1
(
376.5 MHz, CD CN, 208C): d=ꢀ72.8 ppm (d, J(P,F)=706.8 Hz);
3
CID appearance curves were recorded with varying excitation am-
plitudes (0.0–1.0 V), which determined the internal energy scale of
the mass spectrometer (ELAB in V). Relative abundances were calcu-
lated according to Equation (1):
3
1
1
P{ H} NMR (162.0 MHz, CD CN, 208C): d=ꢀ144.6 ppm (hept); ele-
3
mental analysis calcd (%) for C H ClD F N PRu (579.83): C 41.42, H
20
14
8
6
2
3
.88, N 4.83; found: C 41.37, H 4.08, N 4.68%.
Compound 3iBF : The reaction of NaBF (64.7 mg, 0.32 mmol), 3d
P
4
4
ꢀ
ꢁ
fr
i
6
I ðE
Þ
(
37.2 mg, 0.34 mmol), and [{(h -benzene)RuCl } ] (77.0 mg,
fr
tot
i
lab
2
2
P
P
I :ðElabÞ ¼
ð1Þ
fr
i
p
i
0
.15 mmol) in CH Cl (20 mL) gave the product as an orange
i
I
ðElabÞ þ
i
I
ðElabÞ
2
2
powder (32%). M.p. 2428C (decomp); IR (ATR): ~n =3093 (w), 3071
ꢀ
1
1
fr
i
p
i
(
(
w), 1593 (s), 1553 (s), 1048 (vs), 1034 (vs), 781 cm (s); H NMR
in which I =intensity of the fragment ions and I =intensity of the
3
400.1 MHz, CD CN, 208C): d=9.24 (d, J(H,H)=5.56 Hz, 1H; H1),
.62 (d, J(H,H)=4.66 Hz, 1H; H8), 8.15 (m, 2H; H3, H4), 7.72 (td,
J(H,H)=7.98 Hz, J(H,H)=2.72, 1H; H2), 7.46 (d, J(H,H)=4.67 Hz,
parent ions.
3
3
8
Fragmentation amplitude dependent CID spectra were modeled
and fitted by sigmoidal functions of the type shown in Equa-
tion (2)] by using a least-squares criterion.
3
3
3
1
H; H7), 5.84 (s, 6H; Hbenz), 3.55, 3.26 ppm (2ꢂbrs, 6H; H10);
1
3
1
C{ H} NMR (100.6 MHz, CD CN, 208C) d=170.0 (C6), 165.0 (C9),
3
1
59.4 (C8), 156.8 (C1), 155.8 (C5), 141.2 (C3), 129.2 (C2), 125.3 (C4),
ꢀ
ꢁ
1
10.8 (C7), 87.5 ppm (Cbenz), C10 resonances are too broad at RT;
1
50
fr
fit
1
9
11
I ðElabÞ ¼
ð2Þ
F NMR (376.5 MHz, CD CN, 208C): d=ꢀ151.8 ppm; B NMR
1 þ eðElab
ꢀElabÞ B
3
(
128.4 MHz, CD CN, 208C): d=ꢀ1.18 ppm; elemental analysis calcd
3
(
%) for C H ClF N Ru (501.68): C 40.70, H 3.62, N 11.17,; found: C
50
lab
1
7
18B
4
4
The E fit parameter is the amplitude at which the sigmoid func-
tion was at the half-maximum value, whereas B described the rise
4
0.40, H 3.64, N 11.02.
of the sigmoidal curve.
Quantum chemical calculations
+
Quantum chemical calculations on the activation of 3i and the
subsequent formation of the ruthenium(II) hydrido intermediate
Catalysis experiments
[21]
were performed with the Gaussian 03
program by using the
The catalysis experiments were conducted in headspace vials con-
B3LYP gradient-corrected exchange-correlation functional in com-
bination with the 6-31G* basis set for C, H, N, and Cl and the Stutt-
gart/Dresden ECP basis set for Ru. Full geometry optimizations
ꢃ
taining a Teflon -coated magnetic stirring bar, which were predried
ꢀ
3
[22]
in an oven at 608C and then dried under high vacuum (10 mbar)
at 828C for 1 h. After cooling to RT under vacuum, the vial was
were performed in C symmetry by using analytical gradient tech-
1
ꢃ
charged with all solid compounds, sealed with a Teflon -coated
niques and the resulting structures were confirmed to be true
minima by diagonalization of the analytical hessian matrix. The
septum cap (VWR), and the atmosphere in the vial was evacuated
and reflushed with nitrogen three times within 30 min by means
of a syringe. Then isopropanol and other freshly distilled liquids
were added by means of a syringe. Tetradecane or hexadecane
+
starting geometries for the calculations of 3i were taken from
the solid-state structure of compound 3iBF4.
(
50 mL) were used as internal standards for GC analysis. The vials
were placed in an aluminum block kept at 828C by means of
a heating plate equipped with a temperature sensor. At the end of
the reaction, samples were taken by means of a syringe and fil-
X-ray structural analyses
Crystal data and refinement parameters for compounds 3aPF6,
3dBF , 3 fBF , and 3iBF are collected in Table 5. The structures
tered over a short column containing MgSO and silica to remove
4
4
4
4
[23]
[24]
the catalyst. The column was finally washed with a small amount
of ethyl acetate. For GC analysis, a Perkin–Elmer Clarus 580 gas
chromatograph equipped with a flame-ionization detector (FID)
were solved by using direct methods (SHELXS for 3aPF , SIR92
6
for 3dBF , 3 fBF , and 3iBF ), completed by subsequent difference
4
4
4
Fourier syntheses, and refined by full-matrix least-squares proce-
[23]
and a FS-OV-1701-CB-0.25 column (l=30 m, d =0.25 mm; CS Chro-
dures.
(Multiscan) were performed for 3aPF , 3dBF , and 3 fBF ; for 3iBF
4
Semiempirical absorption corrections from equivalents
i
matographie Service GmbH) was used (carrier gas: He, 60 psi; in-
6
4
4
[25]
jector T=2508C; split ratio: 7.1:1; T program: 80!2508C, T ramp:
an analytical numerical absorption correction was applied. All
non-hydrogen atoms were refined with anisotropic displacement
ꢀ
1
5
.678Cmin ).
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
11
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Table 5. Details of crystallographic data, data collection, and refinement.
3
aPF
6
3dBF
4
3 fBF
4
3iBF
4
formula
C
19
H
21Cl
2
F
6
KN
3
PRu
C
21
H
26BClF
4
N
4
Ru
C
19
H
20BCl
2
F
4
N
3
Ru
C
17
H18BClF
4 4
N Ru
M
r
647.43
557.79
549.16
501.68
crystal size [mm]
T [K]
0.45ꢂ0.18ꢂ0.14
150(2)
0.22ꢂ0.14ꢂ0.10
150(2)
0.303ꢂ0.277ꢂ0.058
150(2)
0.516ꢂ0.234ꢂ0.036
150(2)
l [ꢁ]
1.54184
1.54184
0.71073
1.54184
crystal system
space group
a [ꢁ]
b [ꢁ]
c [ꢁ]
orthorhombic
monoclinic
monoclinic
monoclinic
P2
1
2
1
2
1
P2
1
/n
P2
1
/c
1
P2 /n
15.1884(3)
15.4509(3)
10.5274(1)
90
9.4578(1)
10.5033(1)
23.2740(3)
90
14.6441(6)
8.1867(3)
17.8669(7)
90
6.6729(3)
15.3566(7)
18.1079(10)
90
a [8]
b [8]
90
90
98.255(1)
90
2288.04(4)
4
1.619
7.058
106.223(4)
90
2056.71(14)
4
1.774
1.069
99.724(5)
90
1828.91(16)
4
1.822
8.750
g [8]
3
V [ꢁ ]
2470.51(7)
4
1.741
9.794
4.08–62.73
17300
Z
ꢀ
3
1calcd [gcm
]
ꢀ
1
m [mm
q range [ ]
reflns collected
]
o
3.84–62.65
16033
2.879–32.494
25570
3.797–62.837
19585
independent reflns
no. data/restraints/parameters
3945 (Rint =0.0394)
3945/0/301
0.0482, 0.1320
0.0482, 0.1320
1.134
3649 (Rint =0.0229)
3649/0/294
0.0270, 0.0691
0.0280, 0.0696
1.154
6919 (Rint =0.0808)
6919/0/274
0.0779, 0.1972
0.0888, 0.2059
1.215
2937 (Rint =0.0828)
2937/0/255
0.0994, 0.2741
0.1006, 0.2746
1.262
[
a]
final R indices [I>2s(I)]
R indices (all data)
[
b]
GooF
Flack parameter
0.02(2)
–
–
–
ꢀ
3
D1max
/
min [eꢁ
]
2.431/ꢀ0.897
0.965/ꢀ0.575
3.350/ꢀ2.369
2.540/ꢀ1.927
2
2
2
2
1/2
2
2
2
1/2
[a] R1=Sj jF
o
jꢀjF
c
j j/SjF
o
j, wR2=[Sw(F ꢀF ) /SwF ] . [b] GooF=[Sw(F ꢀF ) /(nꢀp)]
.
o
c
o
o
c
parameters. All hydrogen atoms were placed in calculated posi-
tions and refined by using a riding model.
vall, Chem. Eur. J. 2001, 7, 5052–5058; i) J.-E- Bꢄckvall, J. Organomet.
Chem. 2002, 652, 105–111; j) J. Takehara, S. Hashiguchi, A. Fujii, S.-I.
Inoue, T. Ikariya, R. Noyori, Chem. Commun. 1996, 223–224; k) J.-X. Gao,
T. Ikariya, R. Noyori, Organometallics 1996, 15, 1087–1089; l) K.-J. Haack,
S. Hashiguchi, A. Fujii, T. Ikariya, R. Noyori, Angew. Chem. Int. Ed. Engl.
1997, 36, 285–288; m) S. Hashiguchi, A. Fujii, K.-J. Haack, K. Matsumura,
T. Ikariya, R. Noyori, Angew. Chem. Int. Ed. Engl. 1997, 36, 288–290;
Angew. Chem. 1997, 109, 300–303; n) K. Matsumura, S. Hashiguchi, T.
Ikariya, R. Noyori, J. Am. Chem. Soc. 1997, 119, 8738–8739; o) S.-I. Inoue,
K. Nomura, H. Shohei, R. Noyori, Y. Izawa, Chem. Lett. 1997, 26, 957–
CCDC 1498206 (3aPF6), 1498207 (3dBF4) 1498209 (3 fBF4), and
1
498208 (3iBF ) contain the supplementary crystallographic data
4
for this paper. These data are provided free of charge by The Cam-
bridge Crystallographic Data Centre.
Acknowledgements
958; p) R. Noyori, S. Hashiguchi, Acc. Chem. Res. 1997, 30, 97–102; q) K.
Murata, T. Ikariya, R. Noyori, J. Org. Chem. 1999, 64, 2186–2187; r) K.
Murata, K. Okano, M. Miyagi, H. Iwane, R. Noyori, T. Ikariya, Org. Lett.
1999, 1, 1119–1121; s) M. Yamakawa, H. Ito, R. Noyori, J. Am. Chem. Soc.
This study was supported by the German research foundation
DFG within the transregional collaborative research center SFB/
TRR 88 “Cooperative effects in homo and heterometallic com-
plexes” (3MET) and by the state research center OPTIMAS. We fur-
thermore gratefully acknowledge the research college MAGNENZ
and the state research unit NanoKat for financial support.
2
000, 122, 1466–1487; t) R. Noyori, M. Yamakawa, S. Hashiguchi, J. Org.
Chem. 2001, 66, 7931–7944; u) M. Yamakawa, I. Yamada, R. Noyori,
Angew. Chem. Int. Ed. 2001, 40, 2818–2821; Angew. Chem. 2001, 113,
2900–2903; v) I. Yamada, R. Noyori, Org. Lett. 2000, 2, 3425–3427; w) A.
Matsuoka, C. A. Sandoval, M. Uchiyama, R. Noyori, H. Naka, Chem. Asian
J. 2015, 10, 112–115.
[
3] T. Ohkuma, M. Koizumi, K. Muniz, G. Hilt, C. Kabuto, R. Noyori, J. Am.
Chem. Soc. 2002, 124, 6508–6509.
Keywords: density functional calculations · hydrogenation ·
metalation · reaction mechanisms · ruthenium
[4] a) Z.-R. Dong, Y.-Y. Li, J.-S. Chen, B.-Z. Li, Y. Xing, J.-X. Gao, Org. Lett.
005, 7, 1043–1045; b) Z. E. Clarke, P. T. Maragh, T. P. Dasgupta, D. G.
Gusev, A. J. Lough, K. Abdur-Rashid, Organometallics 2006, 25, 4113–
117; c) R. Corberꢆn, E. Peris, Organometallics 2008, 27, 1954–1958;
d) M. Watanabe, Y. Kashiwame, S. Kuwata, T. Ikariya, Eur. J. Inorg. Chem.
012, 504–511.
5] a) P. O. Lagaditis, A. J. Lough, R. H. Morris, J. Am. Chem. Soc. 2011, 133,
662–9665; b) G. Wienhçfer, I. Sorribes, A. Boddien, F. Westerhaus, K.
2
[1] a) H. Meerwein, R. Schmidt, Liebigs Ann. Chem. 1925, 444, 221–233;
b) A. Verley, Bull. Soc. Fr. 1925, 37, 537; c) W. Ponndorf, Angew. Chem.
4
1
926, 39, 138–143; d) R. V. Oppenauer, Recl. Trav. Chim. Pays-Bas 1937,
6, 137.
2] a) R. L. Chowdhury, J.-E. Bꢄckvall, J. Chem. Soc. Chem. Commun. 1991,
2
5
[
[
9
1063–1064; b) N. Menashe, Y. Shvo, Organometallics 1991, 10, 3885–
3891; c) G.-Z. Wang, J.-E. Bꢄckvall, J. Chem. Soc. Chem. Commun. 1992,
337–338; d) M. L. S. Almeida, M. Beller, G.-Z. Wang, J.-E. Bꢄckvall, Chem.
Junge, H. Junge, R. Llusar, M. Beller, J. Am. Chem. Soc. 2011, 133,
12875–12879.
Eur. J. 1996, 2, 1533–1536; e) M. L. S. Almeida, P. Kocovsky, J.-E. Bꢄckvall,
J. Org. Chem. 1996, 61, 6587–6590; f) A. Aranyos, G. Csjernyik, K. J.
Szabo, J.-E. Bꢄckvall, Chem. Commun. 1999, 351–352; g) Y. R. S. Laxmi,
J.-E. Bꢄckvall, Chem. Commun. 2000, 611–612; h) O. Pꢅmies, J.-E. Bꢄck-
[6] R. Castarlenas, M. A. Esteruelas, E. Onate, Organometallics 2008, 27,
3240–3247.
[7] a) M. C. Carrion, F. Sepulveda, F. A. Jalon, B. R. Manzano, Organometallics
2009, 28, 3822–3833; b) A. G. Elliott, A. G. Green, P. L. Diaconescu,
&
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
12
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Dalton Trans. 2012, 41, 7852–7854; c) L.-P. Hea, T. Chena, D.-X. Xuea, M.
Eddaoudia, K.-W. Huang, J. Organomet. Chem. 2012, 700, 202–206; d) J.
DePasquale, M. Kumar, M. Zeller, E. T. Papish, Organometallics 2013, 32,
gin, R. M. Balabin, R. Zenobi, J. Am. Soc. Mass Spectrom. 2010, 21, 172;
f) C. Collette, E. De Pauw, Rapid Commun. Mass Spectrom. 1998, 12,
165–170; g) L. Drahos, R. M. Heeren, C. Collette, E. De Pauw, K. Ve ’key,
J. Mass Spectrom. 1999, 34, 1673–1678.
966–979; e) M. Kumar, J. DePasquale, N. J. White, M. Zeller, E. T. Papish,
Organometallics 2012, 32, 2135–2144.
[18] G. A. Olah, Acc. Chem. Res. 1971, 4, 240–248.
[
[
8] a) E. C. Tyo, A. W. Castleman, Jr., D. Schrçder, P. Milko, J. Roithova, J. M.
Ortega, M. A. Cinellu, F. Cocco, G. Minghetti, J. Am. Chem. Soc. 2009,
[19] S. Hohloch, L. Suntrup, B. Sarkar, Organometallics 2013, 32, 7376–7385.
[20] M. A. Bennett, A. K. Smith, J. Chem. Soc. Dalton Trans. 1974, 233–241.
[21] Gaussian 03, Revision E.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgomery, Jr., T. Vreven,
K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar, J. Tomasi, V. Barone,
B. Mennucci, M. Cossi, G. Scalmani, N. Rega, G. A. Petersson, H. Nakatsu-
ji, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Na-
kajima, Y. Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.
Hratchian, 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. Och-
terski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. Salvador, J. J. Dannenberg,
V. G. Zakrzewski, S. Dapprich, A. D. Daniels, M. C. Strain, O. Farkas, D. K.
Malick, A. D. Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. L. Martin, D. J. Fox, T. Keith, M. A. Al-Laham,
C. Y. Peng, A. Nanayakkara, M. Challacombe, P. M. W. Gill, B. Johnson, W.
Chen, M. W. Wong, C. Gonzalez, J. A. Pople, Gaussian, Inc., Wallingford
CT, 2004.
[22] a) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37, 785; b) A. D. Becke,
Phys. Rev. A 1988, 38, 3098; c) B. Miehlich, A. Savin, H. Stoll, H. Preuss,
Chem. Phys. Lett. 1989, 157, 200; d) P. C. Hariharan, J. A. Pople, Theor.
Chim. Acta 1973, 28, 213; e) D. Andrae, U. Haeussermann, M. Dolg, H.
Stoll, H. Preuss, Theor. Chem. Acc. 1990, 77, 123.
[23] G. M. Sheldrick, Acta Crystallogr. Sect. A 2008, 64, 112.
[24] A. Altomare, G. Cascarano, C. Giacovazzo, A. Guagliardi, M. C. Burla, G.
Polidori, M. Camalli, J. Appl. Crystallogr. 1994, 27, 435.
131, 13009; b) G. Minghetti, S. Stoccoro, M. A. Cinellu, G. L. Petretto, A.
Zucca, Organometallics 2008, 27, 3415.
9] L. Taghizadeh Ghoochany, C. Kerner, S. Farsadpour, F. Menges, Y. Sun, G.
Niedner-Schatteburg, W. R. Thiel, Eur. J. Inorg. Chem. 2013, 4305–4317.
[
[
10] C. Kerner, S.-D. Straub, Y. Sun, W. R. Thiel, Eur. J. Org. Chem. 2016, 3060–
064.
11] a) S. Farsadpour, L. Taghizadeh Ghoochany, Y. Sun, W. R. Thiel, Eur. J.
Inorg. Chem. 2011, 4603; b) S. Farsadpour, L. Taghizadeh Ghoochany, S.
Shylesh, G. Dçrr, A. Seifert, S. Ernst, W. R. Thiel, ChemCatChem 2012, 4,
3
401.
[
12] a) W. R. Thiel, M. Angstl, T. Priermeier, Chem. Ber. 1994, 127, 2373–2377;
b) A.-K. Pleier, H. Glas, M. Grosche, P. Sirsch, W. R. Thiel, Synthesis 2001,
55–62; c) D. Zabel, A. Schubert, G. Wolmershꢄuser, R. L. Jones, Jr., W. R.
Thiel, Eur. J. Inorg. Chem. 2008, 3648–3654; d) H. Brunner, T. Scheck,
Chem. Ber. 1992, 125, 701–709.
[
13] a) D. A. Beauchamp, S. J. Loeb, Chem. Commun. 2002, 2484–2485; b) E.
Bejan, H. A. Haddou, J. C. Daran, G. G. A. Balavoine, Synthesis 1996,
1012–1018; c) E. Ioachim, E. A. Medlycott, M. I. J. Polson, G. S. Hanan,
Eur. J. Org. Chem. 2005, 3775–3780.
[
14] W. R. Browne, C. M. O’Connor, J. S. Killeen, A. L. Guckian, M. Burke, P.
James, M. Burke, J. G. Vos, Inorg. Chem. 2002, 41, 4245–4251.
15] L. Colina-Vegas, W. Villarreal, M. Navarro, C. Rodrigues de Oliveira, A. E.
Graminha, P. I. da S. Maia, V. M. Deflon, A. G. Ferreira, M. R. Cominett,
A. A. Batista, J. Inorg. Biochem. 2015, 153, 150–161.
[
[16] G. Minghetti, S. Stoccoro, M. A. Cinellu, G. L. Petretto, A. Zucca, Organo-
metallics 2008, 27, 3415.
[25] CrysAlisPro, Agilent Technologies, Version 1.171.33.66, 2010, Version
1.171.36.32, 2013 and Version 1.171.37.35, 2014.
[
17] a) V. Gabelica, E. De Pauw, Mass Spectrom. Rev. 2005, 24, 566; b) E.-L.
Zins, C. Pepe, D. Rondeau, S. Rochut, N. Galland, J.-C. Tabet, J. Mass
Spectrom. 2009, 44, 12; c) E.-L. Zins, D. Rondeau, P. Karoyan, C. Fosse, S.
Rochut, C. Pepe, J. Mass Spectrom. 2009, 44, 1668; d) E.-L. Zins, C. Pepe,
D. Schrçder, J. Mass Spectrom. 2010, 45, 1253; e) K. V. Barylyuk, K. Chin-
Manuscript received: October 25, 2016
Final Article published: && &&, 0000
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
13
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

FULL PAPERS
C. Kerner, J. Lang, M. Gaffga, F. S. Menges,
Y. Sun, G. Niedner-Schatteburg,*
W. R. Thiel*
Ready for action! Roll-over cyclometala-
tion occurring at an aminopyrimidinyl-
pyridine ligand activates ruthenium(II)
complexes for base-free transfer hydro-
genation. This process requires a tertiary
amino group attached to the pyrimidine
ring (see figure).
&& – &&
Mechanistic Studies on Ruthenium(II)-
Catalyzed Base-Free Transfer
Hydrogenation Triggered by Roll-Over
Cyclometalation
&
ChemPlusChem 2016, 81, 1 – 14
www.chempluschem.org
14
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!