Aryloxybenzylidene ruthenium chelates: Synthesis, structure and catalytic activity in olefin metathesis

Article abstract of DOI:10.1002/ejic.201301208

New aryloxybenzylidene ruthenium chelates behave like a promising latent catalyst of olefin metathesis. The catalysts are characterised by substantial stability and catalytic inactivity in their dormant forms and a dramatic increase in activity after addi

Full text of DOI:10.1002/ejic.201301208

SHORT COMMUNICATION
DOI:10.1002/ejic.201301208
Aryloxybenzylidene Ruthenium Chelates: Synthesis,
Structure and Catalytic Activity in Olefin Metathesis
[a]
[a]
[a]
˙
Patrycja Zak, Szymon Rogalski, Maciej Kubicki,
Piotr Przybylski,[a] and Cezary Pietraszuk*[a]
Keywords: Metathesis / Ruthenium / Homogeneous catalysis / Chelates / Protonation
New aryloxybenzylidene ruthenium chelates behave like a
promising latent catalyst of olefin metathesis. The catalysts
are characterised by substantial stability and catalytic inac-
tivity in their dormant forms and a dramatic increase in ac-
tivity after addition of a solution of HCl in ether. The mecha-
nism of activation involves protonation of the phenoxide and
the formation of a highly catalytically active hydroxybenzyl-
idene ruthenium chelate.
idene (Figure 2d) ruthenium complexes.[7] In their active
forms, complexes c and d (Figure 2) promote ring-closing
metathesis of diethyl diallylmalonate at a comparable rate,
or even in some cases slightly faster, than parent complexes
(i.e., second-generation Grubbs catalyst and Umicore
M31).[7] In this Communication we report aryloxybenzyl-
idene ruthenium complexes (another representative of
group IV latent catalysts), which are characterized by par-
ticularly high stability in an inactive form and a significant
increase in catalytic activity as a result of chemical acti-
vation. Ruthenium-based olefin metathesis catalysts that
contain an aryloxy ligand are well known.[8] There are also
reports describing ruthenium alkylidenes bearing chelating
aryloxide ligands.[9,10]
Introduction
Olefin metathesis has become a powerful carbon–carbon
bond-formation reaction in organic[1] and polymer chemis-
try.[2] The availability of well-defined ruthenium-based cata-
lysts tolerant of moisture, oxygen, the majority of func-
tional groups and typical processing conditions has made
olefin metathesis widely used in synthesis.[3] Despite dy-
namic progress in the development of olefin metathesis cat-
alysts, many research groups have made efforts to find new
catalysts that exhibit improved catalytic performance and/
or catalysts dedicated to specific applications. One of the
most intensively studied groups of olefin metathesis cata-
lysts is latent catalysts (Figure 1), which, being inactive un-
der standard conditions is characterized by a significant in-
crease in activity as a result of thermal, chemical or photo-
chemical activation.[4]
Figure 2. Group IV latent catalysts of olefin metathesis.
Figure 1. Types of latent catalysts of olefin metathesis.[4b]
Results and Discussion
Known examples of type IV latent catalysts of olefin me-
tathesis include the benzylidenecarboxylate catalyst (Fig-
ure 2a)[5] and recently reported ruthenium nitronate com-
plexes.[6] Recently, we (in cooperation with Grela’s group)
have reported carbamato- (Figure 2b,c) and amidobenzyl-
Treatment of the second-generation Grubbs catalyst (1)
with 2-(prop-1-enyl)phenol (2a; 1 equiv.) in CH2Cl2 at 40 °C
led to a gradual change in colour to green within 24 h.
1
Monitoring of the reaction by H NMR spectroscopy indi-
cates a gradual disappearance of the singlet at δ =
19.07 ppm and the formation of a new singlet at δ =
15.84 ppm in the 1H NMR spectrum. Moreover, the ap-
pearance of a new singlet at δ = 30.33 ppm was observed in
the 31P NMR spectrum. GC–MS analysis of the reaction
mixture confirmed the formation of (prop-1-enyl)benzene
[a] Faculty of Chemistry,
Adam Mickiewicz University in Poznan´,
Umultowska 89b, 61-614 Poznan´
E-mail: pietrasz@amu.edu.pl
http://chemia.amu.edu.pl
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejic.201301208.
Eur. J. Inorg. Chem. 2014, 1131–1136
1131
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
Scheme 1. Synthesis of ruthenium aryloxybenzylidene complexes (3a–d).
and tricyclohexylphosphonium chloride. On the basis of
these observations, the reaction was proposed to proceed
according to Scheme 1.[11]
The synthesis performed in the presence of the equimolar
addition of tricyclohexylphosphine (PCy3), which acted as
the base that bound the HCl liberated during the reaction,
permitted achievement of much higher yields. Our efforts
to synthesize aryloxybenzylidene ruthenium complexes
starting
from
first-generation
Grubbs
catalyst
{[RuCl2(PCy3)2(=CHPh)]} have failed. Complexes 3a–d
were synthesised with isolated yields of 94, 90, 92 and 88%,
respectively, and characterised spectroscopically. They are
stable in air as solids and fairly stable in solution under
argon. Storage of solutions of 3a–d in CD2Cl2 for one week
at room temperature resulted in less than 3% decomposi-
tion in each case. Moreover, prolonged heating of a solution
of complex 3a in C6D6 at 80 °C for 24 h did not lead to
1
more than 5% decomposition (on the basis of H NMR
spectroscopy). The proposed structures of 3a and 3c were
confirmed by X-ray structural analysis (Figure 3).
In both structures the coordination numbers of central
Ru atoms are five, and the coordination is close to square-
pyramidal. Four atoms (Cl, P, O and IMES C) are approxi-
mately coplanar; Ru atoms also lie approximately in this
plane, and are displaced by 0.143(1) Å [0.162(1) Å in 3a]
towards the apical C34 atom {which is 1.968(1) Å
[1.986(1) Å] out of the plane}. That means that the
Ru1=C34 bond is almost perpendicular to the base plane
(it is inclined by approximately 10° from the normal to the
Figure 3. Perspective views of complexes (a) 3a and (b) 3c. The
ellipsoids are drawn at the 33% probability level.
basal plane). The τ parameter,[12] which is a measure of the the dihedral angle between the ring planes is 3.4°), the
distortion between the tetragonal pyramid (ideal value 0) centroid-to-centroid distance is 3.532 Å and the interplanar
and trigonal bipyramid (1), has the value of 0.25 in the case distance is as short as 3.18 Å. In 3a in turn, the potential
of 3c, whereas it is 0.21 for 3a. The double-bond character voids in the crystal structures are occupied by disordered
of Ru1–C34 is clearly seen when comparing its length solvent molecules. The catalytic activity of complexes 3a–d
{1.836(2) Å [1.8499(18) Å in 3a]} to the Ru–C42 bond of has been briefly explored in standard reactions used for
2.071(3) [2.0720(16)]Å. Overall, the introduction of a nitro testing olefin metathesis catalysts.[13] A comparative investi-
substituent in the 4-position of the phenolate ligand does gation has also been performed for the second-generation
not significantly change the coordination of the Ru ion. The Grubbs catalyst (1). The results obtained indicated com-
differences in both bond length and bond angles around plete catalytic inactivity of complexes 3a and 3c and poor
the metal centre are not greater than 3 standard deviations. activity of 3d in ring-opening metathesis polymerization
As expected, the substitution changes the bond pattern in- (ROMP) of 1,5-cyclooctadiene (cod) (0.1 mol-% of 3,
side the ring but does not affect the coordination geometry. CH2Cl2, 40 °C, 0.5 m) (Figure 4). To test the thermoswitch-
In 3c the conformation of the complex can also be deter- ability of catalyst 3a, the ROMP of cod was performed at
mined to some extent by weak intramolecular π···π interac- 80 °C. A slight increase in activity and 8% conversion of
tions between one of the IMES phenyl rings and the ni- monomer was observed after 5 h of the reaction run
trophenyl ring; the rings are almost coplanar (see Figure 3; (0.1 mol-%, benzene, 80 °C, 0.5 m), which demonstrates
Eur. J. Inorg. Chem. 2014, 1131–1136
1132
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
high stability and the nearly complete catalytic inactivity of vated catalysts 3a–d exhibit some catalytic activity in cross-
3a even at elevated temperatures. Our preliminary studies metathesis (CM) of allylbenzene with (Z)-1,4-bis(acetoxy)
on the activation of the aryloxybenzylidenes with HCl indi- but-2-ene (0.5 mol-% of 3, CH2Cl2, 40 °C). Conversion of
cated a strong activation effect of added HCl (solution in 30% and formation of the expected CM product was ob-
ether) on catalytic activity of 3a in ROMP of cod.[14] Com- served after 5 h of the reaction run in the presence of 3b
plete conversion of cod was observed within 1 h using com- (Figure 5).
plex 3a activated by addition of HCl (2 equiv.) (Table 1 and
Figure 4).
Figure 5. CM of allylbenzene with (Z)-1,4-bis(acetoxy)but-2-ene.
Conditions: cat. (0.5 mol-% in relation to allylbenzene), twofold
Figure 4. ROMP of cod. Conditions: cat. (0.1 mol-% in relation to
cod), 0.5 m, CH2Cl2, 40 °C. Squares denote non-activated catalysts,
circles denote catalysts activated with HCl (2 equiv.). Black: 3a;
olive: 3b; magenta: 3c; red: 3d; blue triangles: complex 1 (non-acti-
vated).
excess amount of internal olefin, CH2Cl2, 40 °C. Squares denote
non-activated catalysts; circles denote catalysts activated with HCl
(2 equiv.). Black: 3a; olive: 3b; magenta: 3c; red: 3d; blue triangles:
complex 1 (non-activated).
Addition of a solution of HCl (2 equiv.) in ether leads to
a substantial increase in catalytic activity of all complexes
tested. Conversion of allylbenzene that reached 90% was
observed after 20 min. of the reaction performed in the
presence of activated 3a and 3b, which indicates that both
complexes are attractive catalysts of olefin cross-metathesis
(Figure 5).
To learn more about the origin of activation, complex 3a
was treated in CD2Cl2 at 40 °C with a solution of HCl
(1 equiv.) in ether. A gradual change in colour from dark
green to light green was observed already within 5 min of
Table 1. Activation of catalyst 3a in ROMP of cod.[a]
Entry
Activator
t [h]
Conv. of cod [%]
1
2
3
4
5
none
HCl
F3CCOOH
HBF4
3
1
1
3
1
0
100
9
6
22
CuCl (5 equiv.)
[a] Reaction conditions: complex 3a (0.1 mol-%), activator (2 equiv.
in relation to 3a), 40 °C, CH2Cl2, 0.5 m.
A similar activating effect of the addition of HCl was the reaction course. Monitoring of the reaction by 1H
observed in the presence of complex 3b. In contrast, com- NMR spectroscopy indicates the disappearance of the sing-
plex 3c exhibited poor activity even on prolonged heating let at δ = 15.84 ppm and the formation of a new singlet at
1
at 40 °C in the presence of HCl (Figure 4). Nearly no effect δ = 16.99 ppm in the H NMR spectrum. Moreover, the
on the activity of 3a was observed upon addition of tri- appearance of a new singlet at δ = 32.82 ppm was observed
fluoroacetic or tetrafluoroboric acids (Table 1). Note- in the 31P NMR spectrum. The singlet at δ = 16.99 ppm
worthy, no conversion was observed after 1 h of the reac- (JP,H = 0) assigned to the hydrogen atom at the carbene
tion course upon addition of CuCl (1 equiv.; in relation to carbon suggests that the benzylidene ligand is in the plane
1
catalyst) and conversion of 22% upon addition of a fivefold perpendicular to the Ru–P bond.[15] In the H NMR spec-
excess amount of copper salt. A comparison of the ob- trum no signal was identified that could be assigned to hy-
served limited activating effect of CuCl (acting as phos- drogen from the hydroxy group. However, on the basis of
phine scavenger) and the dramatic effect of addition of HCl the proposed structure of complex 4, it should be expected
(which can not only act as PCy3 scavenger but also as a that coordination of the oxygen atom to ruthenium makes
source of acidic proton able to protonate a Ru–O bond) the proton from the hydroxy group highly acidic and pre-
suggests that protonation of phenolate plays a fundamental sumably highly labile. Indeed, it was found that introduc-
role in the activation of catalyst 3a. Interestingly, inacti- tion of complex 4 to the NMR spectroscopic tube filled
Eur. J. Inorg. Chem. 2014, 1131–1136
1133
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
with CD2Cl2 resulted in an increase in the intensity of the
triplet at δ = 5.31 ppm (assigned to CHDCl2). As we were
1
able to identify all signals in the H NMR spectrum of 4
except the one from the OH group, we believe that fast deu-
teration of the hydroxy group takes place in solution in
CD2Cl2. On the basis of the aforementioned observations,
the reaction of catalyst 3a with HCl was proposed to pro-
ceed according to Scheme 2.
Figure 7. FTIR spectra of 3a (black solid line) and 4 (red dashed
line) in the range from 3100 to 2000 cm–1, recorded in nujol.
Scheme 2. Equimolar reaction of catalyst 3a with HCl.
Structures of catalyst 3a and complex 4 have been ad-
ditionally examined by FTIR spectroscopy and MO-G
PM6 quantum chemical calculations, performed at the sem-
iempirical level of theory. Analysis of FTIR spectra of 3a
and 4 reveals significant differences between them, espe-
cially in the far-infrared region (Figure 6).
Figure 8. The most favoured structures of (a) 3a (ΔHf°
=
–34.49 kcalmol–1) and (b) 4+H2O (ΔHf° = –49.18 kcalmol–1) cal-
culated by the MOG-PM6 method at the semiempirical level of
theory.
Figure 6. FTIR spectra of 3a (black solid line) and 4 (red dashed
line) in the range from 600 to 200 cm–1, the far-infrared region,
recorded in nujol.
The existence of the OH phenolic group in the structure
The calculated frequencies of ν(Ru–Cl) stretching vi-
of 4, in contrast to the structure of 3, can be deduced from brations with the MO-G PM6 method are as follows: a sin-
the FTIR spectrum of the former by the appearance of an gle band at 296 cm–1 for the structure of 3a and two dif-
γ(OH) out-of-plane deformation vibration band with a ferent bands with maxima at 336 and 341 cm–1 for the
maximum at 554 cm–1. Furthermore, in the FTIR spectrum structure of 4+H2O. Taking into account these calculated
of 4, in a higher-frequency region, a broad and low intense values, bands with maxima at 329 and 335 cm–1 of different
band with a maximum at about 2260 cm–1 is observed (Fig- intensities (Figure 6) were assigned to two different types of
ure 7). The appearance of this band corresponds to ν(OH) ν(Ru–Cl) stretching vibrations in the structure of complex
stretching vibrations of the hydrogen-bonded water mole- 4. Despite different substituents in a trans arrangement rela-
cule, as shown in Figure 8, and is evidence of some affinity tive to chlorine atoms, benzylidene and phenolic moieties,
of complex 4 toward water molecules. In the spectrum of the interaction of one of the chlorine ligands with a water
catalyst 3a, the ν(Ru–Cl) stretching vibration band is ob- molecule influenced a very similar position of the two
served at 301 cm–1. Comparison of the FTIR spectra of 3a ν(Ru–Cl) stretching vibration bands. Taking spectroscopic
with 4 (Figure 6) shows that for the latter a complex band and theoretical data into account, in the structure of com-
with a maximum at about 330 cm–1 is observed instead of plex 4, coordination of the oxygen atom of the OH group
one with a maximum at 301 cm–1.
to Ru metal ion, manifested by the presence of a less char-
Eur. J. Inorg. Chem. 2014, 1131–1136
1134
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
acteristic ν(Ru–O) band at 514 cm–1 (Figure 6), increases
the acidity of the proton of this phenolic group.
Experimental Section
Supporting Information (see footnote on the first page of this arti-
cle): General methods and chemicals, syntheses and characteriza-
tion of complexes 3a–d and 4, procedures of catalytic tests.
In turn, such a situation facilitates the formation of the
hydrogen bond between the more acidic OH group and the
oxygen atom of the water molecule, as calculated and
shown in Figure 8. The proton of the phenolic group in the
polarizable hydrogen-bonded system in the six-membered
ring is strongly delocalised between two oxygen atoms and
therefore the ν(O–H) absorption of the phenolic group van-
ishes. PM6 calculations of this structure revealed the pos-
sibility of relatively stable hydrogen bonding of the water
molecule with complex 4 because calculated parameters for
intermolecular hydrogen bonds are: (H2O)O–H···Cl (length
2.6 Å, angle 153°) and (H2O)HO···H–O(phenol) (length
2.8 Å, angle 174°).
Complex 4 was tested in ROMP of cod and was found
to exhibit catalytic activity significantly higher than that of
complex 3a activated with HCl (2 equiv.; Figure 9). This
test clearly demonstrates the importance of activation of the
phenolate ring for the catalytic activity of aryloxybenzyl-
idene complexes.
Acknowledgments
Financial support from the National Science Centre (Poland), (pro-
ject no. UMO-2011/03/B/ST5/01047) is gratefully acknowledged.
[1] For recent reviews on olefin metathesis in organic Synthesis
see: a) J. Cossy, S. Arseniyadis, C. Meyer (Eds.), Metathesis in
Natural Product Synthesis: Strategies, Substrates and Catalysts,
Wiley-VCH, 2010; b) S. P. Nolan, H. Clavier, Chem. Soc. Rev.
2010, 39, 3305–3316; c) R. H. Grubbs, A. G. Wenzel, A. K.
Chatterjee, in Comprehensive Organometallic Chemistry III,
Vol. 11 (Ed.: T. Hiyama), Elsevier, Amsterdam, 2006, pp. 179–
205; d) J. Mulzer, E. Ohler, T. Gaich in Comprehensive Organo-
metallic Chemistry III, Vol. 11 (Ed.: T. Hiyama), Elsevier, Am-
sterdam, 2006, pp. 207–269; e) K. C. Nicolaou, P. G. Bulger,
D. Sarlah, Angew. Chem. 2005, 117, 4564; Angew. Chem. Int.
Ed. 2005, 44, 4490–4527; f) Handbook of Metathesis (Ed.:
R. H. Grubbs), Wiley-VCH, Weinheim, 2003.
[2] For recent reviews on olefin metathesis in polymer chemistry,
see: a) H. Mutlu, L. Montero de Espinosa, M. A. R. Meier,
Chem. Soc. Rev. 2011, 40, 1404–1445; b) K. L. Opper, K. B.
Wagener, J. Polym. Sci., Part A J. Polym. Sci., Part A: Polym.
Chem. 2011, 49, 821–831; c) U. H. F. Bunz, D. Maeker, M.
Porz, Macromol. Rapid Commun. 2012, 33, 886–910; d) S. Sut-
thasupa, M. Shiotsuki, F. Sanda, Polym. J. 2010, 42, 905–915;
e) A. Leitgeb, J. Wappel, C. Slugovc, Polymer 2010, 51, 2927–
2946; f) D. E. Fogg, H. M. Foucault in Comprehensive Organo-
metallic Chemistry III, Vol. 11 (Ed.: T. Hiyama), Elsevier, Am-
sterdam, 2006, pp. 623–652.
[3] For recent reviews on well-defined ruthenium-based catalysts
of olefin metathesis, see: a) F. B. Hamad, T. Suna, S. Xiao, F.
Verpoort, Coord. Chem. Rev. 2013, 257, 2274–2292; b) S. Kress,
S. Blechert, Chem. Soc. Rev. 2012, 41, 4389–4408; c) E. B. An-
derson, M. R. Buchmeiser, Synlett 2012, 23, 185–207; d) G. C.
Vougioukalakis, R. H. Grubbs, Chem. Rev. 2010, 110, 1746–
1787; e) A. M. Lozano-Vila, S. Monsaert, A. Bajek, F. Ver-
poort, Chem. Rev. 2010, 110, 4865–4909; f) C. Samojłowicz, M.
Bieniek, K. Grela, Chem. Rev. 2009, 109, 3708–3742.
[4] For a review on latent catalysts, see: a) Y. Vidavsky, A. Anaby,
N. G. Lemcoff, Dalton Trans. 2012, 41, 32–43; b) A. Szadkow-
ska, K. Grela, Curr. Org. Chem. 2008, 12, 1631–1647; c) S.
Monsaert, A. Lozano Vila, R. Drozdzak, P. Van Der Voort, F.
Verpoort, Chem. Soc. Rev. 2009, 38, 3360–3372.
[5] R. Gawin, A. Makal, K. Wozniak, M. Mauduit, K. Grela, An-
gew. Chem. 2007, 119, 7344; Angew. Chem. Int. Ed. 2007, 46,
7206–7209.
Figure 9. ROMP of cod. Conditions: cat. (0.1 mol-% in relation to
cod), 0.5 m, CH2Cl2, 40 °C. Black squares: complex 3a, blue circles:
complex 3a activated with HCl (2 equiv.), red circles: complex 4.
[6] T. Wdowik, C. Samojłowicz, M. Jawiczuk, M. Malinska, K.
Wozniak, K. Grela, Chem. Commun. 2013, 49, 674–676.
[7] C. Pietraszuk, S. Rogalski, B. Powała, M. Mietkiewski, M. Ku-
bicki, G. Spolnik, W. Danikiewicz, K. Wozniak, A. Pazio, A.
Szadkowska, A. Kozłowska, K. Grela, Chem. Eur. J. 2012, 18,
6465–6469.
Conclusion
[8] a) J. C. Conrad, D. Amoroso, P. Czechura, G. P. A. Yap, D. E.
Fogg, Organometallics 2003, 22, 3634–3636; b) S. Monfette,
D. E. Fogg, Organometallics 2006, 25, 1940–1944; c) J. C. Con-
rad, H. H. Parnas, J. L. Snelgrove, D. E. Fogg, J. Am. Chem.
Soc. 2005, 127, 11882–11883; d) S. Monfette, K. D. Camm,
S. I. Gorelsky, D. E. Fogg, Organometallics 2009, 28, 944–946.
[9] a) M. Jovic, S. Torker, P. Chen, Organometallics 2011, 30, 3971–
3980; b) G. Occhipinti, H.-R. Bjørsvik, K. W. Tornroos, V. R.
Jensen, Organometallics 2007, 26, 5803–5814; c) J. S. M. Samec,
B. K. Keitz, R. H. Grubbs, J. Organomet. Chem. 2010, 695,
1831–1837.
New stable aryloxybenzylidene ruthenium complexes
have been synthesised and characterised. The catalysts are
substantially stable and catalytically inactive (or poorly
active) in their dormant forms. However, a significant in-
crease in activity is observed after “switching on” by ad-
dition of a solution of HCl in ether. A mechanism of acti-
vation that involves the formation of an isolable catalyti-
cally active hydroxybenzylidene complex has been pro-
posed.
Eur. J. Inorg. Chem. 2014, 1131–1136
1135
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.eurjic.org
SHORT COMMUNICATION
[10]
[13] T. Ritter, A. Hejl, A. G. Wenzel, T. W. Funk, R. H. Grubbs,
Organometallics 2006, 25, 5740–5745.
[14] The highest efficiency of activation of complexes 3a–c was ob-
tained when using a twofold molar excess amount of HCl in
relation to the catalyst. A further increase in the concentration
of the acid did not lead to any improvement in catalytic per-
formance.
a) S. Chang, L. R. Jones, C. Wang, L. M. Henling, R. H.
Grubbs, Organometallics 1998, 17, 3460–3465; b) R. Drozdzak,
B. Allaert, N. Ledoux, I. Dragutan, V. Dragutan, F. Verpoort,
Coord. Chem. Rev. 2005, 249, 3055–3074.
[11] When this Communication was under preparation, Samoc and
co-workers described the synthesis and nonlinear optical prop-
erties of complex 3a: C. Wierzbicka, M. Nyk, K. Skowierski,
M. Samoc, Dalton Trans. 2012, 41, 13258–13260.
[12] A. W. Addison, T. N. Rao, J. Reedijk, J. van Rijn, G. C. Ver-
schoor, J. Chem. Soc., Dalton Trans. 1984, 1349–1356.
[15] Compare, for example, E. L. Dias, S. T. Nguyen, R. H. Grubbs,
J. Am. Chem. Soc. 1997, 119, 3887–3897.
Received: September 16, 2013
Published Online: October 29, 2013
Eur. J. Inorg. Chem. 2014, 1131–1136
1136
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim