Formation of different isomers of phosphine-imidazolyl and -pyridyl ruthenium(II) complexes affecting the catalyst activity in the acceptorless dehydrogenation of alcohols

Article abstract of DOI:10.1002/ejic.201402970

The synthesis, reactivity, and catalytic activity of Ru^II complexes with different pyridine- and imidazole-based P,N ligands are reported. The investigations reveal a strong influence of the N-heterocycle and the steric demand of the phosphine groups on the stability of different isomers of [L₂RuX₂] (L = P,N ligand; X = Cl, H). The imidazole-based complex 5 with dicyclohexylphosphine groups was found to be the most active precatalyst for the acceptorless dehydrogenation of primary alcohols, whereas different phosphine groups at the imidazole ligand as well as pyridine-based ligands caused a drop in catalytic activity. In the presence of a primary amine, imines are preferentially formed under these conditions. In summary, the investigations show that comparably small changes in the ligand moiety have a strong effect on the relative stability of stereoisomers for both hydride and chloride complexes, whereas isomerization of the kinetic reaction products was observed in some cases. The described changes in the ligand moiety most probably have a strong impact on the relative stabilities of isomeric intermediates as well and thus affect the catalytic activity of these complexes.

Full text of DOI:10.1002/ejic.201402970

FULL PAPER
DOI:10.1002/ejic.201402970
Formation of Different Isomers of Phosphine–Imidazolyl
and –Pyridyl Ruthenium(II) Complexes Affecting the
Catalyst Activity in the Acceptorless Dehydrogenation of
Alcohols
[a]
[a]
Robert Langer,*[a,b] Alexander Gese, Donatas Gesevicius,
Maximilian Jost,[a] Bastian R. Langer,[a] Felix Schneck,[a]
Alexander Venker,[a] and Weiqin Xu[b]
ˇ
Keywords: Homogeneous catalysis / Dehydrogenation / Ruthenium / N,P ligands / Alcohols
The synthesis, reactivity, and catalytic activity of RuII com- drop in catalytic activity. In the presence of a primary amine,
plexes with different pyridine- and imidazole-based P,N li-
gands are reported. The investigations reveal a strong influ-
ence of the N-heterocycle and the steric demand of the phos-
phine groups on the stability of different isomers of [L2RuX2]
(L = P,N ligand; X = Cl, H). The imidazole-based complex 5
with dicyclohexylphosphine groups was found to be the most
active precatalyst for the acceptorless dehydrogenation of
primary alcohols, whereas different phosphine groups at the
imidazole ligand as well as pyridine-based ligands caused a
imines are preferentially formed under these conditions. In
summary, the investigations show that comparably small
changes in the ligand moiety have a strong effect on the rela-
tive stability of stereoisomers for both hydride and chloride
complexes, whereas isomerization of the kinetic reaction
products was observed in some cases. The described changes
in the ligand moiety most probably have a strong impact on
the relative stabilities of isomeric intermediates as well and
thus affect the catalytic activity of these complexes.
mode of the pincer ligand enforces a trans orientation of
the two hydride ligands in I, which seems to be favored over
the carbonyl ligand trans to the hydride.
Introduction
An important concept in homogeneous catalysis takes
advantage of so-called cooperative ligands or metal–ligand
cooperation.[1] Thereby, the interplay of a coordinated li-
gand with the central metal atom of the complex allows for
the reversible transfer of electrons or protons, thus leading
to the application of base metal complexes as catalysts and
the development of new catalytic reactions.[2]
In this context, a series of pyridine-based pincer com-
plexes have been reported as highly efficient catalysts for
various known and unknown reactions,[3] including the ac-
ceptorless dehydrogenation of primary alcohols in the pres-
ence of amines to amides and the hydrogenation of organic
carbonates to methanol.[4] In particular, the reversible cleav-
age of dihydrogen by an aromatization–dearomatization se-
quence is thought to be responsible for the spontaneous
liberation of H2 in ruthenium pincer complexes
(Scheme 1),[2e] whereas the typical meridional coordination
Scheme 1. Metal–ligand cooperation in H2 activation for different
ruthenium complexes and their structural relationship to P,N-li-
gand systems.
[a] Department of Chemistry, Philipps-Universität Marburg,
Hans-Meerwein-Strasse, 35043 Marburg, Germany
E-mail: robert.langer@chemie.uni-marburg.de
http://www.uni-marburg.de/fb15/ag-langer
[b] Lehn Institute of Functional Materials (LIFM), Sun Yat-Sen
University (SYSU),
In addition to the generation of a reactive trans-dihydride
intermediate, the nature of the donor group D in I seems
to play a crucial role in the significant improvement of the
catalyst activity in acceptorless dehydrogenation reactions
Xingang Road West, Guangzhou 510275, P. R. China
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201402970.
Eur. J. Inorg. Chem. 2015, 696–705
696
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
t
of primary alcohols. Originally it was thought that this MeImCH2PR2 (R = Bu, Ph, p-Tol, Cy) initially results in
group reversibly generates a vacant coordination site, which the formation of [Ru(benzene)(η1-MeImCH2PR2)Cl2],
in many cases is required for the β-hydride elimination of a which isomerizes to the cationic complex [Ru(benzene)(η2-
coordinated alkoxide.[5] But recent experimental studies and MeImCH2PR2)Cl]Cl, thus providing evidence for the hemi-
calculations show that an outer-sphere mechanism is more labile nature of the imidazole ring.[10b] Furthermore, de-
likely.[6] Accordingly, RuII PNN-type pincer complexes ex- pending on the steric demand, phosphine-group mono- and
hibit higher activities in hydrogenation and dehydrogen- bis-ligated complexes were obtained by the reaction of
ation reactions than their PNP-type analogues.[3,4] In line [Ru(cod)(methylallyl)2] (cod = cyclooctadiene) with two
with these findings, a similar activity is observed for RuII equivalents of imidazolyl phosphine and HBr.[10c] Herein,
complexes with a tetradentate phenanthroline-based ligand we present the synthesis and characterization of RuII com-
II, which retains the trans orientation of the two ancillary plexes with the general composition [L2RuCl2] (Scheme 2),
ligands. In contrast to the pyridine-based pincer system, the with L denoting a pyridyl- or imidazolyl-based P,N ligand.
intermediately formed trans-dihydride complex II un- Furthermore, we compare all complexes in the acceptorless
dergoes reversible hydrogenation of a C=C double bond of dehydrogenation of primary alcohols and demonstrate for
the ligand backbone instead of H2 liberation.[7]
the most active catalyst that the imine is formed preferably
As the trans arrangement of the phosphine groups does in the presence of a primary amine. Finally, we investigated
not seem to be crucial for the performance of the com- the impact of the heterocycle and the phosphine group by
plexes, the formal deduction to two P,N ligands should also comparing the reactivity and structure of the in situ gener-
lead to active dehydrogenation catalysts with the formal ated dihydride complexes.
composition [L2RuX2] (X = H, Cl, vacant; L = P,N ligand).
In addition, these bidentate P,N ligands are an important
class of ligands that has been widely used in homogeneous
Results and Discussion
catalysis.[8] Furthermore, octahedral complexes that contain
two of these P,N ligands can potentially exist as different
Treatment of [(Ph3P)3RuCl2] with diisopropylphos-
stereo- and constitutional isomers, which, for example, has phino(2-picolyl)phosphine (PicPiPr2; 2 equiv.) in dichloro-
been shown to be relevant for hydride complexes that can methane or THF results in the formation of three different
serve as active hydrogenation catalysts.[9]
complexes according to the 31P{1H} NMR spectrum of the
For this reason, we started to investigate the coordina- reaction mixture, which shows two sharp singlet resonances
tion behavior of pyridyl- and imidazolyl-functionalized of different intensity at δ = 73.4 and 75.3 ppm, as well as
phosphine ligands towards common ruthenium precursors. two doublet resonances centered at δ = 68.9 and 70.4 ppm
Although hydrogenation reactions that employ in situ gen- (2JP,P = 29.7 Hz). Similar results were obtained when [(cod)-
erated ruthenium catalysts with picolyl- and methylimid- RuCl2]n was employed instead, thereby suggesting that no
azolyl (MeIm) phosphines have been reported, the reverse PPh3 is present in the three newly formed complexes. On
reaction has not been investigated so far.[10] However, the the basis of 1H-COSY, 1H-gNOESY, 1H,13C-HSQC, and
reaction of [{Ru(benzene)Cl2}2] with one equivalent of 1H,13C-HMBC NMR spectroscopic experiments as well as
Scheme 2. Synthesis of complexes 1–5.
Eur. J. Inorg. Chem. 2015, 696–705
697
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
selective 31P-decoupling NMR spectroscopic experiments,
With the aim of varying the steric and electronic proper-
assignments for the three complexes could be made, which ties of the P,N-ligand system, we employed the analogous
suggested the molecular structure for 1a–c shown in 2,6-lutidine-based ligand 2,6LutPiPr2 in the complexation of
Scheme 2. Interestingly, no chemical exchange could be de- [(Ph3P)3RuCl2]. For the corresponding 2,6LutPPh2 ligand,
tected between the three species in the mixture by 1H- Lavigne and Lugan et al. reported the formation of the
gNOESY NMR spectroscopy, but the ratio of 1a, 1b, and trans-dichlorido complex, whereas with 2,6LutPCy2 only
1c slowly changes over the course of 20 h in CD2Cl2 accord- one equivalent of ligand reacted with the RuII precursor.[12]
ing to the 31P{1H} NMR spectra.[11] All attempts to sepa- For the iPr-substituted ligand it turned out that, despite the
rate these complexes by crystallization failed and resulted increased steric repulsion of the methyl groups in the 6-
in the formation of different types of crystals and the pre- position, the trans-dichlorido complex 2 was formed. In
cipitation of a yellow powder. Nevertheless, it was possible analogy to 1a, the central RuII atom in 2 exhibits a dis-
to determine the molecular structure for two of the three torted-octahedral environment with two coplanar 2,6Lut-
complexes by means of single-crystal X-ray diffraction, PiPr2 ligands (Є = 9.44°) and two chlorido ligands in the
thereby confirming that different [(PicPiPr2)2RuCl2] isomers apical positions. In comparison to 1a, the Ru–P bond
are formed in this reaction (Figure 1a and b). The bond lengths in 2 are slightly shorter (Table 1), whereas the Ru–
lengths and angles of the structurally characterized com- N distances are elongated in 2, possibly owing to the steri-
plexes are summarized in Table 1.
cally more demanding methyl substituents. With a value of
In the symmetric trans-dichloride complex 1a the Cl1– 53.62°, the tilt angle between the two pyridine groups in 2
Ru1–Cl2 angle was found to be 172.94°, whereas the two is slightly larger than in complex 1 (45.50°).
coordinated PicPiPr2 ligands in the octahedral RuII com-
The 31P{1H} NMR spectrum of 2 in C6D6 exhibits a
plex exhibit an almost coplanar orientation with the two single sharp peak at δ = 71.7 ppm. Contrary to these obser-
i
PiPr2 groups cis to each other. Comparably long Ru–P dis- vations, very broad resonances are observed for the Pr
tances between 2.310 and 2.312 Å are observed in complex groups and the benzylic CH2 groups, whereas the reso-
1a, whereas Ru–N bond lengths range between 2.154 and nances that correspond to the pyridine ring and the CH3
2.160 Å. The second complex was identified as the corre- group in the 6-position of the pyridine ring appear rather
1
1
sponding cis-dichlorido complex 1b with the two pyridine sharp in the H NMR spectrum. The H-gNOESY NMR
rings located trans to each other (Figure 1b). In comparison spectrum of 2 in C6D6 displays chemical exchange between
i
to 1a, the Ru–N and Ru–P distances are slightly shorter, the resonances of the Pr groups, respectively, as well as be-
whereas the bond length of the ruthenium–chlorine bond tween the two broad resonances that correspond to the
1
appears longer in 1b.
benzylic CH2 protons at δ = 3.20 and 5.12 ppm in the H
Figure 1. Molecular structures of 1a, 1b, and 2 obtained by single-crystal diffraction. (a) ORTEP diagram of 1a; (b) ORTEP diagram of
1b; and (c) ORTEP diagram of 2. All thermal ellipsoids are set at 50% probability (hydrogen atoms were omitted for clarity).
Table 1. Selected bond lengths [Å] and angles [°] of compounds 1–5 for comparison.
1a
1b
2
3
4
5
Ru–P
Ru–Cl
Ru–N
Cl–Ru–Cl
N–Ru–N
P–Ru–P
PLig–Ru–NLig
2.310–2.312
2.432–2.442
2.154–2.160
172.94
90.72
111.63
2.285
2.509
2.077
87.94
173.67
104.77
82.68
43.38
2.270–2.274
2.423–2.449
2.278–2.286
176.04
98.80
104.64
2.262–2.269
2.484–2.500
2.049–2.108
87.16–87.35
172.8–175.2
104.74–105.38
80.88–81.52
45.47–45.77
2.249–2.273
2.459–2.482
2.074–2.132
90.92
90.47
104.14
2.281–2.294
2.482–2.530
2.078–2.096
80.93–81.75
174.5–174.6
103.53–105.29
80.1–81.1
78.77–79.92
78.57–78.59
53.62
80.50–81.00
61.44
Het1–Ru–Het2 45.50
38.82–39.38
Eur. J. Inorg. Chem. 2015, 696–705
698
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
NMR spectrum. Such observations are in accordance with the preparation of this manuscript, exhibits the same stereo-
the fluxionality phosphine arms, which are able to adopt chemical configuration as complex 4.[10c] With values of
different orientations.
2.249–2.273 Å, the Ru–P bond lengths were found to be
To further investigate the impact of the aromatic substit- almost identical with the recently reported dibromido com-
uent on the structure, the reactivity, and the catalytic ac- plex. In addition, the similarity between the bromido and
tivity, we synthesized imidazolyl-based P,N ligands with a the chlorido complex is reflected in identical Ru–N bond
more electron-rich N-heterocycle bound to the phosphino lengths, whereas the ruthenium–halogen bond exhibits typi-
groups. In addition to late-transition-metal complexes with cal lengths for each element, respectively.
imidazolyl phosphines,[13] Beller and co-workers charac-
terized a series of complexes with the general composition cyclohexylphosphine-substituted ligand MeImCH2PCy2 ex-
[Ru(benzene)(η2-MeImCH2PR2)Cl]Cl.[10b]
hibits similar electron-donating properties, whereas the
In comparison to MeImCH2PiPr2, the corresponding di-
The analogous iPr-substituted ligand MeImCH2PiPr2 bulky cyclohexyl groups cause increased steric demand.
readily reacts with [(Ph3P)3RuCl2] (0.5 equiv.) to the cis- Utilizing MeImCH2PCy2 in the complexation of
dichlorido complex 3, in which the phosphine groups are [(Ph3P)3RuCl2] in THF at 60 °C resulted in the formation
located trans to the two chlorido ligands (Figure 2a). The of a yellow precipitate in 55% yield, which was identified
octahedral complex 3 crystallizes as a mixture of the Λ and as the dicationic dimer [(MeImCH2PCy2)4Ru2Cl2]Cl2 (5).
Δ isomer in the asymmetric unit of the triclinic unit cell. Interestingly, when two equivalents of MeImCH2PCy2 were
The Ru–P distances in 3 are slightly longer than in 1a, 1b, utilized in the reaction with [Ru(cod)(methylallyl)2] and
and 2, whereas the bonds to the nitrogen atoms of the imid- HBr, only one equivalent of ligand binds to the central
azole ring are found to be shorter than in the pyridine ana- ruthenium atom.[10b]
logues. Importantly, the same isomer as with PicPiPr2 seems
The structural analysis of 5 revealed a dimeric complex
to be the thermodynamic product of the complexation reac- with two octahedrally coordinated RuII atom, which are
tion and no other isomers are formed, whereas the methyl bridged by two μ2-coordinating chlorido ligands (Fig-
substituent in the 6-position of 2,6LutPiPr2 apparently fa- ure 2c). Similar to 1b, the phosphine groups are located
vors the trans-dichlorido complex. Although the resonances trans to the chlorido ligands, thereby causing shortening of
of the iPr groups appear slightly broadened in the 1H NMR the Ru–P bond (2.281–2.294 Å) relative to 1a and 2. With
spectrum of 3, no chemical exchange could be detected in values between 2.482–2.530 Å, the Ru–Cl bond lengths to
1
the H-gNOESY NMR spectrum. In accordance with the the bridging chlorido ligands are very similar to the dis-
determined structure, the 31P{1H} NMR spectrum of 3 ex- tances found in the other complexes. The imidazole rings in
hibits a singlet resonance at δ = 81.1 ppm.
5 occupy the axial positions in this dimer and are arranged
As small changes in the ligand moiety lead to a changed in a pairwise and almost parallel manner with short dis-
coordination behavior and different reaction products, we tances of 3.355–3.626 Å, thus indicating π interactions be-
started to investigate the influence of the steric and elec- tween the imidazole rings.
tronic properties of the phosphine arm. Accordingly, the
Owing to the similarity of the ligand scaffold to other
sterically less demanding and more π-accepting phosphine pyridine- and phenanthroline-containing ruthenium com-
MeImCH2PPh2 was utilized in the reaction with [(Ph3P)3- plexes, we investigated complexes 1–5 as precatalysts for the
RuCl2], leading to the chiral cis-dichlorido complex 4. The dehydrogenation of primary alcohols. Accordingly, in a typ-
molecular structure of 4 has been confirmed by single-crys- ical experiment,
a
ruthenium dichlorido complex
tal X-ray diffraction (Figure 2b); it shows a chiral octahe- (0.01 mmol), KOtBu (0.04 mmol), and primary alcohol
dral complex with two chlorido ligands cis to each other, (5.00 mmol) in toluene (4 mL) were heated to reflux
but with different groups in the trans position. The corre- (Table 2). As the reaction proceeds rather slowly, relatively
sponding bromido complex, which was published during long reaction times were necessary to obtain high produc-
Figure 2. Molecular structures of 3–5 obtained by single-crystal diffraction. (a) ORTEP diagram of 3; (b) ORTEP diagram of 4; and (c)
ORTEP diagram of 5. All thermal ellipsoids are set at 50% probability (hydrogen atoms were omitted for clarity).
Eur. J. Inorg. Chem. 2015, 696–705
699
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
Table 2. Acceptorless dehydrogenation of primary alcohols.[a]
Entry
Catalyst
Alcohol
S/B/C[b]
t [h]
Conversion[c] [%]
Yield ester[c] [%] Yield hemi-acetal[c] [%]
1
2
3
1
benzyl alcohol
benzyl alcohol
benzyl alcohol
benzyl alcohol
benzyl alcohol
benzyl alcohol
n-hexanol
n-hexanol
n-hexanol
n-hexanol
n-hexanol
500:4:1
500:4:1
500:4:1
500:4:1
1000:4:1
500:4:1
500:4:1
500:4:1
500:4:1
1000:4:1
500:4:1
200:4:1
400:4:1
120
120
120
120
120
120
120
120
120
120
120
19
63.5
41.8
70.3
76.1
95.0
36.8
41.5
71.1
41.6
93.7
43.7
27.4
93.2
38.8
3.3
61.7
56.2
63.2
2.4
37.6
45.0
30.9
68.2
0.4
16.8
35.4
24.3
29.3
4.4
11.4
31.5
32.8
2.4
25.4
4.5
19.2
6.7
2
3
4
4
5[d]
6
5
[(Ph3P)3RuCl2]
7
8
9
1
3
4
10[d]
11
12[e]
13[e]
5
[(Ph3P)3RuCl2]
4
5
benzyl alcohol
benzyl alcohol
1.2
46.5
19
[a] Reaction conditions: Alcohol (5.00 mmol), KOtBu (0.04 mmol), catalyst (0.01 mmol) and m-xylene (internal standard; 1.00 mmol),
and toluene (4 mL) were heated to reflux for 120 h. [b] Ratio of substrate/base/catalyst. [c] Determined by GC analysis with m-xylene as
internal standard. [d] Catalyst (0.05 mmol) was used. [e] Precatalyst (4: 0.01 mmol, 5: 0.005 mmol) and KOtBu (0.04 mmol) were added
under air; substrates were added under argon without purification and heated to reflux for 19 h.
tivity of the catalyst and to address selectivity issues. Em- is similar to 3 and 4 but the hemi-acetal is formed in much
ploying complexes 1a–c as catalyst in the dehydrogenation higher yield. For comparison we utilized the non-function-
of benzyl alcohol gave the corresponding ester in 38.8% alized precursor complex [(Ph3P)3RuCl2] as catalyst for the
and the hemi-acetal in 24.3% yield (Table 2, entry 1), as dehydrogenation of benzyl alcohol (Table 2, entry 6). To
detected by GC analysis. Interestingly, after removal of the our surprise, the reaction resulted in 36.8% conversion of
solvent and the internal standard, only benzaldehyde, benzyl alcohol, but the hemi-acetal is formed almost exclu-
benzyl benzoate, and benzyl alcohol could be detected as sively. Thus, complex 2 and [(Ph3P)3RuCl2] exhibit similar
reaction products. The hemi-acetal seems stable under the catalytic activities and selectivities in the acceptorless de-
conditions of the GC and its formation is likely base-cata- hydrogenation of benzyl alcohol. In accordance with the
lyzed. In a control experiment, a mixture of benzaldehyde discussed hemi-acetal formation, the catalytic dehydrogen-
and benzyl alcohol in toluene was heated to reflux in the ation of a mixture of benzaldehyde and benzyl alcohol with
presence of KOtBu; it showed a slow formation of the hemi- complex 5 as catalyst results in the conversion of both sub-
acetal. It is worth mentioning that homogeneous catalysts strates and subsequent ester formation. Moreover, the con-
capable of dehydrogenating a primary alcohol to an alde- tinuous GC analysis of the reaction with complex 5 as cata-
hyde are rare.[14]
lyst revealed that after a short period the concentration of
Interestingly, the methyl-substituted complex 2 exhibits a aldehyde and hemi-acetal remained constant over the
significantly lower catalytic productivity in the dehydroge- course of the reaction.
nation reaction, in which only 3.3% of benzyl benzoate and
The reaction with n-hexanol as substrate resulted in
29.3% of benzaldehyde hemi-acetal are formed (Table 2, en- much lower conversion of 41.5% with 1 as catalyst but gave
try 2). With a conversion of 70.3%, the imidazolyl-based a similar yield of the corresponding ester (Table 2, entry 7).
complex cis-[(MeImCH2PiPr2)2RuCl2] (3) exhibits a produc- When the imidazole-based complex 3 was employed in the
tivity similar to its pyridine-based counterpart (1a–c), but reaction, 71.1% conversion of n-hexanol was observed, but
the selectivity for the ester formation is increased. A similar only 45.0% of hexyl hexanoate was formed during this reac-
observation can be made for the phenyl-substituted ana- tion (Table 2, entry 8). The phenyl-substituted complex 4
logue 4, which results in a conversion of 76.1% when em- appears less active for the dehydrogenation of n-hexanol
ployed as a catalyst in the dehydrogenation of benzyl and gave only 41.6% conversion (Table 2, entry 9). In line
alcohol (Table 2, entry 4). In comparison to 3, with 4 as with the observations for 3, 93.7% conversion of n-hexanol
catalyst slightly less of the ester and more of the hemi-acetal was detected with complex 5 as a catalyst (Table 2, entry
is formed. The highest catalytic activity was observed with 10), whereas the ester was formed in 68.3% and the corre-
complex 5 as catalyst, which resulted in 95% conversion of sponding hemi-acetal in 19.2% yield. Finally, we investi-
the benzyl alcohol (Table 2, entry 5). When the reaction was gated the activity of the precursor complex [(Ph3P)3RuCl2]
stopped after 66 h, 69.0% conversion of benzyl alcohol was as catalyst, which resulted in 43.7% conversion of n-hexa-
detected by means of GC analysis. Thus, complex 5 exhibits nol, but the major product of the reaction turned out to be
a similar catalytic activity to the previously reported com- 2-butyl octanol (Table 2, entry 11), which is likely formed
plex II with a four-dentate and phenanthroline-based li- by aldol condensation of hexanal followed by hydrogena-
gand.[8] Importantly, at 63.2%, the yield of benzyl benzoate tion of the formed α,β-unsaturated aldehyde.
Eur. J. Inorg. Chem. 2015, 696–705
700
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
As the imidazole-based dichlorido ruthenium complexes active complex of this series, complex 5, as a dehydrogen-
did not show any sign of decomposition in the presence of ation catalyst in the presence of primary amine
a
oxygen and moisture after two weeks, we investigated the (Scheme 4). Notably, the conversion of benzyl alcohol de-
possibility of using these complexes as bench-stable precata- creased in the presence of benzylamine but resulted in selec-
lysts under air. In a typical experiment, complex 4 or 5, tive formation of the imine (39.0%). In contrast to the im-
which had been stored under air for several days, was ine formation, which is assumed to take place in a noncata-
weighted into a Schlenk tube together with the required lytic dehydration step of the generated aldehyde and the
amount of KOtBu. The tube was evacuated and refilled primary amine, the generation of amides or esters requires
with argon and benzyl alcohol, then m-xylene and toluene an additional dehydrogenation step of the in situ formed
were added to the Schlenk tube under a stream of argon hemi-aminal or the hemi-acetal, respectively.[3c,21] Accord-
without any kind of purification or degassing of these com- ingly, with higher catalyst loading (0.2 mol-%), complete
pounds. With 27.4% conversion of the primary alcohol, conversion of benzyl alcohol was observed, but a very unse-
complex 4 showed poor performance after heating the mix- lective product distribution with almost equal amounts of
ture to reflux for 19 h (Table 2, entry 12). To our surprise, imine, ester, and amide was obtained. Notably, almost no
with 0.25 mol-% of 5 as precatalyst, 93.2% conversion of secondary amine (Ͻ0.5%) was observed in these reactions.
benzyl alcohol was observed after 19 h (Table 2, entry 13).
In addition to determining the catalytic activity of the
Interestingly, the corresponding hemi-acetal was the main dichlorido complexes, we aimed to compare the reactivity
product of this reaction, whereas benzyl benzoate was and stability of the corresponding dihydride complexes with
formed in only 35.4% yield.
the analogous complexes that contained three- and four-
In the presence of primary amines, different products dentate ligands. Several mechanistic investigations of dif-
have been reported for different catalysts in the acceptorless ferent ruthenium catalysts for the acceptorless dehydroge-
dehydrogenation (Scheme 3).[15] These range from N-het- nation of primary alcohols and related reactions have been
erocyclic compounds[16] to secondary amines,[17] imines,[18] conducted, thus indicating that trans-dihydridoruthen-
and amides.[4a,19,20] For this reason we investigated the most ium(II) complexes are usually intermediates in the catalytic
Scheme 3. Acceptorless dehydrogenation of benzyl alcohol in the presence of benzylamine.
Scheme 4. Formation of hydride complexes.
Eur. J. Inorg. Chem. 2015, 696–705
701
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
cycle.[22] For this reason, we treated the synthesized dimensional NMR spectroscopy, the cis-dihydride isomer
dichlorido complexes with two equivalents of a hydride with the two phosphine groups in a trans position to them
transfer reagent, such as NaHBEt3 in C6D6. It turned out is assumed for both complexes. The two other products
that the generated hydride complexes under discussion were from the reaction of 3 with two equivalents of NaHBEt3
not isolable as pure compounds. In spite of this, a possible were identified as the chiral cis-dihydride 7b and a trans-
structure of the formed complexes was suggested on the dihydride complex (7c), with the assumed structure shown
basis of 1H, 1H{31P}, and 1H-gCOSY NMR spectros- in Scheme 4.
copy.
When 1a–c were treated with two equivalents of
NaHBEt3 (0.5 m in C6D6), the solution immediately decolo-
rized and the 31P{1H} NMR spectrum indicated the forma-
tion of a new ruthenium complex, which was identified as
the cis-dihydride complex 6a. In solution, complex 6a is
slowly converted to another hydride complex, the NMR
spectra of which are in agreement with the formation of
complex 6b. Selected NMR spectroscopic chemical shifts
for the prepared hydride complexes are summarized in
Table 3.
Table 3. Selected NMR chemical shifts for the hydride complexes
6–8.
31P{1H} NMR
[ppm]
1H NMR
[ppm]
2JP,H [Hz][a]
2JH,H
[Hz][a]
6a 67.9 (d), 96.5 (d)
(2JP,P = 14.8 Hz)
6b 88.4 (s)
–16.83 (vtd)
–7.52 (ddd)
–14.77 (t)
–7.68 (m)
–17.95 (vtd)
–7.76 (ddd)
–9.14 (t)
21.9
91.4, 25.6
20.1
–
32.9
97.5, 24.4
17.1
8.5
8.5
–
Figure 3. 1H NMR spectra of the dihydride complexes 6–8 in
C6D6: (a) 1a–c + NaHBEt3 (2 equiv.) after 5 min; (b) 1a–c +
NaHBEt3 (2 equiv.) after 18 h; (c) 3 + NaHBEt3 (2 equiv.); and
(d) 5 + NaHBEt3 (2 equiv.).
7a 72.7 (s)
–
7b 76.4 (d), 103.9 (d)
(2JP,P = 14.8 Hz)
7c 82.5 (s)
8.5
8.5
–
However, in all cases the in situ formed dihydride com-
plexes 6–8 appeared stable in solution for several days with-
out any sign of dihydrogen elimination, but they are con-
verted into a mixture of unidentified products after workup,
including removal of all volatiles under vacuum. Further-
8
64.2 (s)
–7.79 (m)
–
[a] Coupling constant of the resonance corresponding to the hy-
dride ligand.
The isomerization of ruthenium hydride complexes was more, the addition of a primary alcohol such as benzyl
shown to be relevant for ruthenium diamine diphosphine alcohol or methanol to these complexes resulted in immedi-
complexes, in which the corresponding trans-dihydride ate gas evolution of dihydrogen and the formation of a new
complex was found to be the active species, and which exhi- complex, respectively. In the case of complex 8, the newly
bits a limited lifetime and isomerizes to the corresponding formed complex after addition of benzyl alcohol gave rise
cis-dihydride complexes.[9] This is of particular importance to a broadened resonance at δ = 78.6 ppm in the 31P{1H}
1
because for some ruthenium catalysts the cis and the trans NMR spectrum. In the corresponding H NMR spectrum,
isomer of hydride complexes are calculated to have similar two broad resonances at δ = –10.57 and –1.17 ppm were
energy but exhibit significantly different reaction barriers observed. Notably, heating of this mixture to 70 °C for one
within the catalytic cycle.[22b]
hour resulted in the evolution of a new resonance at δ =
Whereas with the pyridine-based P,N ligand the chiral 85.9 ppm in the 31P{1H} NMR spectrum and a new reso-
complex 6a is initially formed as the kinetic product, which nance at δ = –21.77 ppm for the corresponding hydride li-
1
slowly converts into the symmetric hydride complex 6b, the gand in the H NMR spectrum. Addition of CD3OD to
situation changed for the imidazole-based ligands. Treat- the dihydride complexes caused gas evolution as well and
ment of complex 3 with NaHBEt3 (2 equiv.) in C6D6 re- resulted in the formation of the analogous complexes, but
sulted in the formation of a mixture of three isomeric dihy- no deuterium incorporation into the P,N ligand could be
dride complexes (7a–c), whereas the ratio of these com- observed after several days. In contrast to the reactivity of
plexes did not change for several weeks. In contrast, the the RuII complexes with rigid three- and four-dentate li-
same reaction with complex 5 led to a single product (8), gands, in which a trans-dihydride complex or a reaction
as indicated by a singlet resonance at δ = 64.2 ppm in the product thereof can be observed, the more flexible biden-
31P{1H} NMR spectrum. Complexes 7a and 8 show very tate P,N ligands allow for the formation of different cis-
similar patterns in the 1H NMR spectrum for the resonance dihydride complexes, which could generally be formed by
that corresponds to the imidazole moiety and the hydride isomerization of an initially formed trans-dihydride com-
ligands (Figure 3c and d). On the basis of one- and two- plex.
Eur. J. Inorg. Chem. 2015, 696–705
702
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
For comparison we investigated the reaction mixture of tivity of the synthesized precatalysts. We could demonstrate
a catalytic dehydrogenation experiment by NMR spec- in part that kinetic reaction products slowly isomerize to
troscopy. Complex 5 was treated with KOtBu (2 equiv.) and thermodynamically stable isomers at ambient temperature.
benzyl alcohol (50 equiv.) in toluene or C6D6. Initially, a Overall, the relative stabilities of isomeric intermediates are
new complex with a chemical shift of δ = 80.4 ppm in the likely influenced by the changes in the ligand moiety and
31P{1H} NMR spectrum was formed. The absence of a thus cause a different catalytic performance.
hydride resonance in the 1H NMR spectrum might indicate
the formation of a symmetric alkoxide complex. Upon heat-
Experimental Section
ing to at least 70 °C, a new complex was formed within
minutes with a chemical shift of δ = 83.2 ppm in the
31P{1H} NMR spectrum. The newly formed complex exhib-
its a broad resonance at δ = –22.60 ppm with an integral of
General: All experiments were carried out under an atmosphere of
purified argon or nitrogen in a Braun Labmaster glovebox or by
using standard Schlenk techniques. CH2Cl2 was dried with CaH2,
one in its 1H NMR spectrum, which corresponded to a toluene was dried with sodium, and THF was dried with Na/K
alloy. C6D6 was distilled from Na/K alloy and stored over molec-
ular sieves; all other deuterated solvents were sparged with argon
and stored over molecular sieves. [(Ph3P)3RuCl2],[23] [(cod)-
RuCl2]n,[24] and the P,N ligands were prepared according to pre-
viously reported procedures.[12,13d] Benzyl alcohol and n-hexanol
were purchased from Aldrich, degassed, and stored in the glovebox
under an inert atmosphere. Detailed procedures for the synthesis
of all complexes including all spectroscopic data can be found in
the Supporting Information.
hydride ligand. The chemical shift of δ = –22.60 ppm is
quite similar to those observed after the addition of benzyl
alcohol to complex 8 and indicates the presence of a weak
donor ligand, such as an alcohol or an alkoxide, trans to
the hydride. Notably, prolonged heating of the mixture did
not change this situation. That no dihydride complexes can
be observed in the presence of alcohols and the fact that
the separately prepared dihydride complexes readily liberate
H2 upon treatment with alcohol indicates that the observed
hydride alkoxide species likely represents the catalytic rest-
ing state. The experiments with CD3OD showed that no
deuterium incorporation into the P,N ligand takes place,
thus suggesting that none of the ligand protons is involved
in hydrogen generation and that an intermediate dihydrogen
complex is probably formed by simple protonation of the
dihydride.
As the comparably small changes in the ligand moiety
have a strong effect on the relative stability of stereoisomers,
for both hydride and chloride complexes, it seems likely that
the relative stabilities of possible intermediates (on and off
cycle) are also strongly affected by these changes and thus
cause different catalytic performance. Whereas complexes
with rigid ligand systems are rather limited in their possible
stereoisomers, the use of flexible P,N ligands has led to dif-
ferent isomeric species that involve isomerization equilibria.
1H, 13C, and 31P NMR spectra were recorded with a Bruker DRX
1
400 or a DPX 250 NMR spectrometer. H and 13C{1H}, 13C-APT
(attached proton test) NMR spectroscopic chemical shifts are re-
ported in ppm downfield from tetramethylsilane. The resonance of
the residual protons in the deuterated solvent was used as internal
1
standard for H NMR spectroscopy. The solvent peak of the deu-
terated solvent was used as internal standard for 13C NMR spec-
troscopy. 31P NMR spectroscopic chemical shifts are reported in
ppm downfield from H3PO4 and referenced to an external 85%
solution of phosphoric acid in D2O. IR spectra were recorded by
attenuated total reflection of the solid samples with a Bruker Ten-
sor IF37 spectrometer. HR-ESI mass spectra were acquired with a
LTQ-FT mass spectrometer (Thermo Fischer Scientific), as were
HR-APCI mass spectra. In both cases the resolution was set to
100000. Elemental analyses were performed with a Vario Micro
Cube Elemental Analyzer. Gas chromatography was performed
with m-xylene as internal standard with an HP-5 column and an
Agilent 6850 series GC system.
Single-Crystal X-ray Analysis: The single-crystal X-ray diffraction
data for the structural analysis of 1a, 1b·THF, 2·0.5THF,
3·2.5CH2Cl2, 4·3CH2Cl2, and 5·3CHCl3 were collected using
graphite-monochromated Mo-Kα radiation (λ = 0.71073 Å) with a
STOE IPDS2 or IPDS2T imaging plate detector system. The struc-
tures were solved by direct methods with SHELXS-97 and refined
against F2 by full-matrix least-square techniques using SHELXL-
97.[25] Numerical absorption corrections were applied on the basis
of the crystal descriptions.[26]
Conclusion
In conclusion, we have described the synthesis and reac-
tivity of RuII complexes with different P,N ligands, thereby
demonstrating the influence of the N-heterocycle as well as
the influence of the steric and electronic properties of the
phosphine on the relative stability of different isomers of
the type [L2RuX2] (L = P,N ligand; X = Cl, H). Among
the investigated complexes, the bench-stable precatalyst 5
exhibits the highest catalytic activity in the dehydrogenation
of primary alcohols to esters and is able to catalyze the
direct transformation of alcohols to imines in the presence
of amines. The reactivity studies with the corresponding
hydride complexes provide evidence for a non-cooperative
mechanism in which a hydrido alkoxide species appears to
be the catalytic resting state. In addition, our investigations
show that small changes in the ligand moiety strongly affect
CCDC-994528 (for 1a), -994529 (for 1b·THF), -994530 (for
2·0.5THF), -994531 (for 3·2.5CH2Cl2), -994532 (for 4·3CH2Cl2),
and -994533 (for 5·3CHCl3) contain the supplementary crystallo-
graphic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
Dehydrogenation of Primary Alcohols: In a typical experiment, cat-
alyst (0.01 mmol), KOtBu (0.04 mmol), primary alcohol
(5.00 mmol), m-xylene (1.00 mmol), and toluene (4 mL ) were
placed in a Schlenk tube under nitrogen atmosphere. In the case of
the dimeric complex 5, 0.005 mmol was used in the catalytic reac-
the stability of different isomers as well as the catalytic ac- tion. The mixture was heated to reflux for the specified time, and
Eur. J. Inorg. Chem. 2015, 696–705
703
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
[11]
[12]
For details, please see the Supporting Information.
E. Mothes, S. Sentets, M. A. Luquin, R. Mathieu, N. Lugan,
G. Lavigne, Organometallics 2008, 27, 1193–1206.
the extent of conversion was frequently checked by GC analysis.
After the reaction, the mixture was cooled to ambient temperature
and the mixture was analyzed by GC.
[13]
[14]
a) L. D. Field, B. A. Messerle, K. Q. Vuonga, P. Turner, Dalton
Trans. 2009, 3599–3614; b) H.-F. Lang, P. E. Fanwick, R. A.
Walton, Inorg. Chim. Acta 2002, 329, 9–12; c) M. A. Jalil, S.
Fujinami, H. Nishikawa, J. Chem. Soc., Dalton Trans. 2001,
1091–1098; d) M. A. Jalil, S. Fujinami, H. Senda, H. Nishi-
kawa, J. Chem. Soc., Dalton Trans. 1999, 1655–1661.
a) R. Kawahara, K. Fujita, R. Yamaguchi, Angew. Chem. Int.
Ed. 2012, 51, 12790–12794; b) K. Fujita, T. Yoshida, Y. Imori,
R. Yamaguchi, Org. Lett. 2011, 13, 2278–2281; c) R. Kawa-
hara, K. Fujita, R. Yamaguchi, J. Am. Chem. Soc. 2012, 134,
3643–3646.
a) G. E. Dobereiner, R. H. Crabtree, Chem. Rev. 2010, 110,
681–703; b) G. Guillena, D. J. Ramón, M. Yus, Chem. Rev.
2010, 110, 1611–1641.
a) S. Michlik, R. Kempe, Angew. Chem. Int. Ed. 2013, 52,
6326–6329; b) S. Michlik, R. Kempe, Nat. Chem. 2013, 5, 140–
144; c) D. Srimani, Y. Ben-David, D. Milstein, Angew. Chem.
Int. Ed. 2013, 52, 4012–4015.
a) F. E. Fernández, M. C. Puerta, P. Valerga, Organometallics
2012, 31, 6868–6879; b) D. Hollmann, A. Tillack, D. Michalik,
R. Jackstell, M. Beller, Chem. Asian J. 2007, 2, 403–410; c) A.
Del Zotto, W. Baratta, M. Sandri, G. Verardo, P. Rigo, Eur. J.
Inorg. Chem. 2004, 524–529; d) B. Blank, M. Madalska, R.
Kempe, Adv. Synth. Catal. 2008, 350, 749–758; e) M. Haniti,
S. A. Hamid, J. M. J. Williams, Chem. Commun. 2007, 725–727;
f) R. Grigg, T. R. B. Mitchell, S. Sutthivaiyakit, N. Tongpeny,
J. Chem. Soc., Chem. Commun. 1981, 611–612; g) M. H. S. A.
Hamid, C. L. Allen, G. W. Lamb, A. C. Maxwell, H. C. May-
tum, A. J. A. Watson, J. M. J. Williams, J. Am. Chem. Soc.
2009, 131, 1766–1774; h) Y. Watanabe, Y. Tsuji, H. Ige, Y.
Ohsugi, T. Ohta, J. Org. Chem. 1984, 49, 3359–3363; i) Y. Wat-
anabe, Y. Morisaki, T. Kondo, T. Mitsudo, J. Org. Chem. 1996,
61, 4214–4218; j) D. Gnanamgari, E. L. O. Sauer, N. D. Schley,
C. Butler, C. D. Incarvito, R. H. Crabtree, Organometallics
2009, 28, 321–325.
a) N. D. Schley, G. E. Dobereiner, R. H. Crabtree, Organome-
tallics 2011, 30, 4174–4179; b) C. Chen, Y. Zhang, S. H. Hong,
J. Org. Chem. 2011, 76, 10005–10010; c) A. Maggi, R. Madsen,
Organometallics 2012, 31, 451–455; d) M. A. Esteruelas, N.
Honczek, M. Oliván, E. Oñate, M. Valencia, Organometallics
2011, 30, 2468–2471; e) Y. Nakajima, Y. Okamoto, Y.-H.
Chang, F. Ozawa, Organometallics 2013, 32, 2918–2925; f) C.
Xu, L. Y. Goh, S. A. Pullarkat, Organometallics 2011, 30,
6499–6502.
a) L. U. Nordstrøm, H. Vogt, R. Madsen, J. Am. Chem. Soc.
2008, 130, 17672–17673; b) J. H. Dam, G. Osztrovszky, L. U.
Nordstrøm, R. Madsen, Chem. Eur. J. 2010, 16, 6820–6827; c)
M. H. G. Prechtl, K. Wobser, N. Theyssen, Y. Ben-David, D.
Milstein, W. Leitner, Catal. Sci. Technol. 2012, 2, 2039–2042.
a) S. C. Ghosh, S. Muthaiah, Y. Zhang, X. Xu, S. H. Hong,
Adv. Synth. Catal. 2009, 351, 2643–2649; b) S. C. Ghosh, S. H.
Hong, Eur. J. Org. Chem. 2010, 4266–4270; c) S. Muthaiah,
S. C. Ghosh, J.-E. Jee, C. Chen, J. Zhang, S. H. Hong, J. Org.
Chem. 2010, 75, 3002–3006; d) Y. Zhang, C. Chen, S. C.
Ghosh, Y. Li, S. H. Hong, Organometallics 2010, 29, 1374–
1378; e) C. Chen, S. H. Hong, Org. Biomol. Chem. 2011, 9, 20–
26; f) J. Zhang, M. Senthilkumar, S. C. Ghosh, S. H. Hong,
Angew. Chem. Int. Ed. 2010, 49, 6391–6393.
Supporting Information (see footnote on the first page of this arti-
cle): Details of the data collection and the refinement are described
in the Supporting Information.
Acknowledgments
R. L. is grateful to Prof. S. Dehnen and Prof. C. v. Hänisch for
their continuous support and assistance. This research is supported
by the Erich Becker Foundation. W. X. received a PhD fellowship
from the National Natural Science Foundation of China (NSFC).
[15]
[16]
[1] a) H. Grützmacher, Angew. Chem. Int. Ed. 2008, 47, 1814–
1818; Angew. Chem. 2008, 120, 1838–1842; b) S. Schneider, J.
Meiners, B. Askevold, Eur. J. Inorg. Chem. 2012, 3, 412–429;
c) J. I. van der Vlugt, Eur. J. Inorg. Chem. 2012, 3, 363–375; d)
D. Milstein, Top. Catal. 2010, 53, 915–923; e) C. Gunanathan,
D. Milstein, Science 2013, 341, 249–261.
[2] a) S. Schneider, J. Meiners, B. Askevold, Eur. J. Inorg. Chem.
2012, 412–429; b) D. Gelman, S. Musa, ACS Catal. 2012, 2,
2456–2466; c) S. Kuwata, T. Ikariya, Chem. Commun. 2014, 50,
14290–14300; d) V. T. Annibale, D. Song, RSC Adv. 2013, 3,
11432–11449; e) C. Gunanathan, D. Milstein, Acc. Chem. Res.
2011, 44, 588–602.
[3] a) J. Zhang, G. Leitus, Y. Ben-David, D. Milstein, J. Am. Chem.
Soc. 2005, 127, 10840–10840; b) E. Balaraman, B. Gnanaprak-
asam, L. J. W. Shimon, D. Milstein, J. Am. Chem. Soc. 2010,
132, 16756–16758; c) B. Gnanaprakasam, J. Zhang, D.
Milstein, Angew. Chem. Int. Ed. 2010, 49, 1468–1471; Angew.
Chem. 2010, 122, 1510–1513; d) C. Gunanathan, L. J. W. Shi-
mon, D. Milstein, J. Am. Chem. Soc. 2009, 131, 3146–3147; e)
B. Gnanaprakasam, D. Milstein, J. Am. Chem. Soc. 2011, 133,
1682–1685; f) B. Gnanaprakasam, E. Balaraman, Y. Ben-Da-
vid, D. Milstein, Angew. Chem. Int. Ed. 2011, 50, 12240–12244;
Angew. Chem. 2011, 123, 12448–12452; g) C. Gunanathan, D.
Milstein, Angew. Chem. Int. Ed. 2008, 47, 8661–8664; Angew.
Chem. 2008, 120, 8789.
[4] a) C. Gunanathan, Y. Ben-David, D. Milstein, Science 2007,
317, 790–792; b) E. Balaraman, C. Gunanathan, J. Zhang,
L. J. W. Shimon, D. Milstein, Nat. Chem. 2011, 3, 609–614.
[5] a) O. Blum, D. Milstein, J. Organomet. Chem. 2000, 593–594,
479–484; b) C. M. Fafard, O. V. Ozerov, Inorg. Chim. Acta
2007, 360, 286–292; c) N. A. Smythe, K. A. Grice, B. S. Wil-
liams, K. I. Goldberg, Organometallics 2009, 28, 277–288; d)
J. C. M. Ritter, R. G. Bergman, J. Am. Chem. Soc. 1998, 120,
6826–6827.
[17]
[18]
[19]
[20]
[6] a) M. Montag, J. Zhang, D. Milstein, J. Am. Chem. Soc. 2012,
134, 10325–10328; b) S. Qu, Y. Dang, C. Song, M. Wen, K.-
W. Huang, J. Am. Chem. Soc. 2014, 136, 4974–4991; c) H. Li,
X. Wang, F. Huang, G. Lu, J. Jiang, Z.-X. Wang, Organometal-
lics 2011, 30, 5233–5247; d) H. Li, X. Wang, M. Wen, Z.-X.
Wang, Eur. J. Inorg. Chem. 2012, 5011–5020.
[7] R. Langer, I. Fuchs, M. Vogt, E. Balaraman, Y. Diskin-Posner,
G. Leitus, L. J. W. Shimon, Y. Ben-David, D. Milstein, Chem.
Eur. J. 2013, 19, 3407–3414.
[8] M. P. Carrol, P. J. Guiry, Chem. Soc. Rev. 2014, 43, 819–833,
and references cited therein.
[21]
[22]
H. Li, X. Wang, M. Wen, Z.-X. Wang, Eur. J. Inorg. Chem.
2012, 5011–5020.
[9] R. Abbel, K. Abdur-Rashid, M. Faatz, A. Hadzovic, A. J.
Lough, R. H. Morris, J. Am. Chem. Soc. 2005, 127, 1870–1882.
[10] a) T. Miura, I. E. Held, S. Oishi, M. Naruto, S. Saito, Tetrahe-
dron Lett. 2013, 54, 2674–2678; b) K. Junge, B. Wendt, F. A.
Westerhaus, A. Spannenberg, H. Jiao, M. Beller, Chem. Eur. J.
2012, 18, 9011–9018; c) S. Werkmeister, K. Junge, B. Wendt,
A. Spannenberg, H. Jiao, C. Bornschein, M. Beller, Chem. Eur.
J. 2014, 20, 4227–4231.
a) X. Yang, ACS Catal. 2013, 3, 2684–2688; b) I. S. Makarov,
P. Fristrup, R. Madsen, Chem. Eur. J. 2012, 18, 15683–15692;
c) M. Montag, J. Zhang, D. Milstein, J. Am. Chem. Soc. 2012,
134, 10325–10328; d) D. Cho, K. Chul Ko, J. Y. Lee, Organo-
metallics 2013, 32, 4571–4576; e) K. S. Sandhya, C. H. Suresh,
Organometallics 2013, 32, 2926–2933.
Eur. J. Inorg. Chem. 2015, 696–705
704
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
FULL PAPER
[23]
[26] a) X-SHAPE, v. 1.06, Crystal Optimisation for Numerical Ab-
sorption Correction Program, Stoe & Cie GmbH, Darmstadt,
Germany, 1999; b) X-RED32, v. 1.01, Data Reduction Pro-
gram, Stoe & Cie GmbH, Darmstadt, Germany, 2001.
Received: October 09, 2014
P. S. Hallman, T. A. Stephenson, G. Wilkinson, Inorg. Synth.
1970, 12, 237–240.
[24] M. O. Albers, E. Singleton, J. E. Yates, Inorg. Synth. 1989, 26,
249–258.
[25] G. M. Sheldrick, Acta Crystallogr., Sect. A 2008, 64, 112–122.
Published Online: January 19, 2015
Eur. J. Inorg. Chem. 2015, 696–705
705
© 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim