Synthesis and comparative behavior of ruthena(II)cycles bearing benzene ligand in the radical polymerization of styrene and vinyl acetate

Article abstract of DOI:10.1016/j.jorganchem.2015.10.013

A series of half-sandwich ruthenium(II) complexes of the type [Ru(η⁶-C₆H₆)(N∩C)L]PF₆ (L = PPh₃, P(n-Bu)₃, SbPh₃, MeCN), bearing cyclometalated N,N-dimethylbenzylamine (1a-d) and

Full text of DOI:10.1016/j.jorganchem.2015.10.013

Journal of Organometallic Chemistry 799-800 (2015) 299e310
Contents lists available at ScienceDirect
Journal of Organometallic Chemistry
journal homepage: www.elsevier.com/locate/jorganchem
Synthesis and comparative behavior of ruthena(II)cycles bearing
benzene ligand in the radical polymerization of styrene and vinyl
acetate
,
b
a
a
Vanessa Martínez Cornejo a , Jessica Olvera Mancilla , Salvador Lopez Morales ,
ꢀ
b
b
a, **
ꢀ
ꢀ
ꢀ
Jose Alberto Oviedo Fortino , Simon Hernandez-Ortega , Larissa Alexandrova
,
,
*
Ronan Le Lagadec b
a
ꢀ
Instituto de Investigaciones en Materiales, UNAM, Circuito Exterior s/n, Ciudad Universitaria, 04510 Mexico D. F., Mexico
Instituto de Química, UNAM, Circuito Exterior s/n, Ciudad Universitaria, 04510 Mexico D. F., Mexico
b
ꢀ
a r t i c l e i n f o
a b s t r a c t
6-C6H6)(N∩C)L]PF6 (L ¼ PPh3, P(n-
Article history:
A series of half-sandwich ruthenium(II) complexes of the type [Ru(
h
Received 30 August 2015
Received in revised form
5 October 2015
Accepted 8 October 2015
Available online 22 October 2015
Bu)3, SbPh3, MeCN), bearing cyclometalated N,N-dimethylbenzylamine (1aed) and 2-phenylpyridine (2a
ed) moieties, has been efficiently prepared by ligand substitution. The molecular structures of the new
compounds were unequivocally determined by single-crystal X-ray diffraction. The catalytic activity of
the complexes in the radical polymerizations of styrene and vinyl acetate was evaluated, and a
comparative structure e activity analysis was performed.
© 2015 Elsevier B.V. All rights reserved.
Keywords:
Cyclometalated ruthenium complexes
Ligand exchange
Atom transfer radical polymerization
1. Introduction
understood, even though ruthenacycles have shown remarkable
photophysical and electrochemical properties. For instance, they
Cyclometalation of ligands by transition metals is one of the
easiest ways to form organometallic compounds with a metal-
carbon sigma bond. Due to the stabilization by chelation, these
compounds are generally quite robust and therefore may easily be
managed without extreme precautions. The platinum group metals
are by far the most popular domain used for cyclometalation re-
actions and a vast number of metallocycles has been prepared by
heteroatom-assisted CeH bond activation, with the palladium
complexes being the most studied, since cyclopalladated de-
rivatives are known for nearly all classes of ligands [1e4]. The
corresponding ruthenium compounds have been much less
investigated and their synthesis and properties are less well
have been considered as promising materials for applications in
solar cells, intervalence electron-transfer systems or in bio-
electronic devices as efficient electron shuttles for oxidoreductase
enzymes [1,2,5e11]. The organoruthenium derivatives also possess
the required characteristics for the interaction with biomolecules
and several ruthenacycles are considered as promising candidates
for anticancer drugs [1,12e14]. Although ruthenium complexes
have been widely investigated for homogeneous catalysis, their
ortho-metalated counterparts have not as yet demonstrated their
full potential in this area and only a few examples of cyclo-
ruthenated compounds showing good activity in homogeneous
catalysis have been reported [15e17].
Cycloruthenation is very versatile and of broad scope mostly
because of the great diversity and availability of ruthenium pre-
cursors. However, one setback is the relative lack of reactivity of
ruthenium complexes toward the cyclometalation reaction and
frequently the synthetic route requires various steps and results in
low yields [1,18]. Our group has been studying the synthesis and
possible ways of application of cycloruthenated compounds for
almost two decades. Simple and highly effective synthesis of
ruthenacycles has been developed which allows easy modification
Abbreviations: BEB, 1-(bromoethyl)-benzene; MMA, methyl methacrylate;
St, styrene; VAc, vinyl acetate; MEK, methyl ethyl ketone; TEMPO, (2,2,6,6-
tetramethyl-piperidin-1-yl)oxyl; Cyclometalated ligands: dmba, N,N-dime-
thylbenzylamine; phpy, 2-phenylpyridine; tolpy, 2-(p-tolyl)pyridine.
* Corresponding author.
** Corresponding author.
E-mail addresses: laz@unam.mx (L. Alexandrova), ronan@unam.mx (R. Le
Lagadec).
http://dx.doi.org/10.1016/j.jorganchem.2015.10.013
0022-328X/© 2015 Elsevier B.V. All rights reserved.

300
V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
of the compounds in the desirable direction [8,9,19e21].
labile ligands in the catalytic activity, a series of ten cyclometalated
ruthenium(II) complexes bearing
6-C6H6 has been prepared (See
One of the promising applications of the ruthenacycles in
catalysis may be in atom transfer radical polymerization (ATRP) or
metal catalyzed living radical polymerization. Indeed, ruth-
enium(II) complexes were among the first catalysts reported for
this reaction and to the date remain one of the most active and
versatile catalytic systems [22,23]. Moreover, ruthenium complexes
have been known to be among the most efficient for CeC bond
formation in the Karasch addition or atom transfer radical addition
(ATRA), a reaction mechanistically very similar to ATRP [24e26].
The classical mechanism of ATRP consists in the reversible ho-
molysis of a terminal carbon-halogen bond of the dormant species
through the abstraction of the halogen atom by the metallic catalyst
resulting in the formation of growing radicals and the complex
in þ1 higher oxidation state (Scheme 1). The complex participates
in a reversible oxidative addition reaction and therefore its catalytic
activity should correlate with its redox potential, where a lower
redox potential means higher catalytic activity. A direct correlation
has been shown for structurally simpler copper catalysts [27], while
the structure e ATRP reactivity relationship for the ruthenium-
based catalysts is highly sophisticated and no simple dependence
has been found, even for structurally similar ruthenium com-
pounds [28e30]. Cyclometalated ruthenium(II) complexes may
have advantages here in comparison with their coordinated con-
geners because of the less positive RuIII/RuII potential [1,12,31e33].
Several ruthenacycles have been reported as efficient catalysts in
ATRP and ATRA reactions of different acrylic and vinylic compounds
[26,30,34e36]. They may be better referred to catalyst precursors
since all of them are 18 electrons coordinatively saturated mole-
cules and a vacant site must be generated to enable their activation
for ATRP [30,34e36]. Recently, the very active catalyst precursor
h
Fig. 1 for the structures). In order to evaluate the influence of
different parameters such as the lability of the ligands and the
electron density on the metal centre, the structure of the complexes
was gradually modified: (i) two different cycloruthenated moieties,
N,N-dimethylbenzylamine (group 1) and 2-phenylpyridine (group
2) were introduced, and, (ii) a substitution reaction allowed the
incorporation of various ancillary ligands such as acetonitrile,
phosphines and stibines. Such a variety of compounds enabled the
investigation of the influence of the different ligands on the cata-
lytic activity and to clarify some mechanistic aspects. The behavior
of the complexes was analyzed under conditions of the polymeri-
zation of St and VAc. Furthermore, considering that the ionic
character of the complexes may also impact on their catalytic
performance [26,38,39], two neutral [Ru(h
6-C6H6)(N∩C)Cl] com-
pounds (group 3), with a phpy-based cyclometalated fragment,
were also studied and the results compared with group 1 and 2
catalysts.
2. Experimental
2.1. Materials and reagents
All reactions were carried out under inert atmosphere (dini-
trogen or argon) using conventional Schlenk glassware; all solvents
were dried using established procedures and distilled under dini-
trogen prior to use. Styrene (99%) was washed three times with
1 wt % NaOH solution and passed through a column filled with
neutral alumina, vinyl acetate (>99%, Aldrich) was passed through a
neutral alumina column, distilled under reduced pressure, and
stored under nitrogen. All the others reagents were purchased from
Aldrich and used as received: N,N-dimethylbenzylamine, 2-
phenylpyridine, n-decane, BEB (97%), carbon tetrachloride, tetra-
hydrofuran HPLC, tri-n-butylphosphine, triphenylphosphine,
triphenylstibine, potassium hexafluorophosphate, silver hexa-
fluorophosphate, anhydrous ether and anhydrous acetonitrile.
Commercial RuCl3 was purchased from Pressure Chemical Com-
pany. Complexes 1d, 3′, 300 and 4 were prepared according to the
literature [18,19,40e42].
[Ru(h
6-C6H6)(dmba)(MeCN)]PF6 has been reported, which was able
to polymerize vinyl acetate, one of the most difficult monomers for
ATRP, via the reversible activation of the carbon-Cl terminals under
specific conditions [37]. Additionally, the investigation demon-
strated the importance of the presence of a benzene ligand in the
structure of the catalyst, since the complexes having benzene were
generally more active than their counterparts with polypyridine
ligands. Unfortunately, polymerization of other more conjugated
monomers, such as styrene or methyl methacrylate proceeded
without control.
Mass Spectra were obtained using a JEOL JMS-SX102A instru-
ment with m-nitrobenzyl alcohol as the matrix [FABþ mode, m/z (%,
To continue our effort in understanding the role of potentially
Scheme 1. Mechanism of ATRP.

V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
301
Fig. 1. Synthetic routes, structures of ruthenacycles and numbering scheme for NMR assignation used in the present study.
relative abundance) throughout]. Infrared spectra were recorded
on an Alpha ATR spectrometer from Bruker. 1H (300 MHz) and 31
{1H} (121.6 MHz) NMR spectra were recorded using a JEOL GX in-
strument. The scale is used throughout; chemical shifts are in
ppm, and the coupling constants are in Hz. Elemental analyses were
obtained on an Exeter Analytical CE-440.
Conversions were determined from the concentration of resid-
ual monomer measured by GC using a Shimadzu GC-2010 gas
chromatograph equipped with one capillary column RESTEK sta-
were acquired and processed using Shimadzu GC-MS solution
software.
P
d
2.2. Crystallography
Crystalline yellow prisms for 1ae1c, and 2ae2d were grown
independently from CH2Cl2/diethyl ether and mounted on glass
fibers. In all cases, the X-ray intensity data were measured at 298 K
on a Bruker SMART APEX CCD-based X-ray diffractometer system
bilwax (30 m, 0.53 mm ID, and 0.5 mmdf) with n-decane as an in-
equipped with a Mo-target X-ray tube (
was placed at a distance of 5.0 cm from the crystals in all cases. A
total of 1800 frames were collected with a scan width of 0.3 in
l
¼ 0.71073 Å). The detector
ternal standard in every polymerization. Analysis conditions were
as follows: injector temperature, 220 ꢀC; temperature program,
4 min 40 ꢀC, 15 ꢀC/min until 220 ꢀC, 2 min 220 ꢀC. Molecular
weights (Mn) and molecular weight distributions (Mw/Mn) of the
polymers were determined using a gel permeation chromatograph
Waters 2695 ALLIANCE Separation Module, equipped with two HSP
gel columns in series (HR MBL molecular weight range from 5102 to
7105 and MB-B from 103 to 4106) and a RI Waters 2414 detector.
THF was used as an eluent at 35 ꢀC with a flow rate of 0.5 mL/min.
Linear polystyrene standards were used for the calibration. Theo-
retical molecular weights were calculated without taking into ac-
u
and an exposure time of 10 s/frame. The frames were integrated
with the Bruker SAINT Software package using a narrow-frame
integration algorithm [43]. The integration of the data was done
using a triclinic unit cell for 1b, orthorhombic for 2d and mono-
clinic for 1a, 1c, 2a, 2b, 2c, to yield a total of 24735 for 1a, 13676 for
1b, 25448 for 1c,125407 for 2a, 25286 for 2b, 35321 for 2c and
23295 for 2d reflections to a maximum 2
q
angle of 50.00, of which
0.0324], 5785
5529 [R(int) 0.0579], 6006 [R(int)
¼
¼
[R(int) ¼ 0.0797], for 1a e 1c, 5678 [R(int) ¼ 0.0645], 5696
[R(int) ¼ 0.0898], 5973 [R(int) ¼ 0.0940], 3804 [R(int) ¼ 0.0583] for
2a e 2d were independent. Analysis of the data showed in all cases
negligible decays during data collections. The structures were
solved by Patterson method using SHELXS-2012 program [44]. The
remaining atoms were located via a few cycles of least squares
refinements and difference Fourier maps. Hydrogen atoms were
input at calculated positions, and allowed to ride on the atoms to
which they are attached. Thermal parameters were refined for
hydrogen atoms on the phenyl groups using a Ueq ¼ 1.2 Å to pre-
cedent atom in all cases. For all complexes, the final cycle of
refinement was carried out on all non-zero data using SHELXL-
count the end groups according to the following equation: Mn,th
¼
([Monomer]0/[Initiator]0)
0 ꢂ conversion ꢂ1.
ꢁ
Conversion
ꢁ
MWmonomer, where
GC-MS analysis were performed on a Shimadzu GC-2010/MS-
QP2010s system equipped with an AOC-20i auto sampler, with
injector temperature of 320 ꢀC, a 1:5 split ratio and injection vol-
ume of 1 mL. Capillary column separation was a 0.25 mm thick film
[30 m x 0.32 mmID Rtx®-5MS (RESTEK) with a 5 m integra-guard
column] using a flow rate of 1.21 mL/min and 70 kPa helium
pressure. The chromatograms were acquired in the electron impact
scan mode at 70 eV with a mass range of 40e900 (m/z). The data

302
V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
2014/7 [44]. The PF6 anion in all compounds, the CH2Cl2 solvent in
compound 1b and the arene ring in compounds 2b and 2d are
disordered and were refined in two major contributors and refined
anisotropically.
The solution was passed through a short column of Al2O3 using
CH2Cl2 as an eluent. A pale yellow fraction was collected and
evaporated to dryness. Crystallization from CH2Cl2/diethyl ether
(slow diffusion) gave yellow crystals, which were washed with
diethyl ether and dried under vacuum.
2.3. General procedure for complexes 1
2.4.1. Synthesis of 2a
In a typical experiment, a solution of 1d (200 mg, 0.40 mmol)
and 0.6 mmol of the ligand (1a triphenylphosphine, 1b tri-n-
butylphosphine or 1c triphenylstibine) in 30 mL of MeOH was
stirred for 20 h at 45 ꢀC. The solvent was evaporated under vacuum,
and the residue was dissolved in CH2Cl2. The solution was passed
through a short column of Al2O3 using CH2Cl2 as an eluent. A pale
yellow fraction was collected and evaporated to dryness. Crystal-
lization from CH2Cl2/diethyl ether (slow diffusion) gave yellow
crystals, which were washed with diethyl ether and dried under
vacuum.
2.4.1.1. Yield. 341 mg, 85%. 1H NMR (CD3COCD3): 9.26 (d, 1H,
3JHH ¼ 5.5, H8), 7.81 (d, 1H, 3JHH ¼ 7.5, H1), 7.53 (t, 1H, 3JHH ¼ 7.7 Hz,
3
H2), 7.47e7.38 (m, 2H, H3 and H6), 7.32 (dd, 3H, JHH ¼ 7.9 Hz,
4JHH ¼ 6.1 Hz, H4, H5 and H7), 7.25e7.14 (m, 6H, PPh3), 7.01e6.77
(m, 9H, PPh3), 6.03 (s, 6H, C6H6). 31P{1H} NMR (CD3COCD3): 43.84
(s, PPh3), ꢃ144.19 (stp, 1JPF ¼ 707.8 Hz, PF6). M/S FABþ: 596 [(M þ H)
e PF6] (82%), 518 [(M þ H) e (C6H6 þ PF6)] (38%), 334 [(M þ H) e
(PPh3 þ PF6)] (18%), 256 [(M þ H) e (C6H6 þ PPh3 þ PF6)] (8%). IR
(FTIR): y 832 (PF6, s). Anal.: Calcd.: C35H29F6NP2Ru$CH2Cl2: C, 52.38,
H, 3.78; N, 1.70. Found: C, 52.98; H, 3.82; N ¼ 1.83.1
2.3.1. Synthesis of 1a
2.4.2. Synthesis of 2b
2.3.1.1. Yield. 202 mg, 70%. 1H NMR (CD3CN): 8.04 (d, 1H,
3JHH ¼ 8 Hz, H1), 7.61e7.30 (m, 15H, PPh3), 7.16 (t, 1H, 3JHH ¼ 7.2 Hz,
H2), 6.98 (t, 1H, 3JHH ¼ 7 Hz, H3), 6.63 (d, 1H, 3JHH ¼ 7.5, H4), 5.60 (s,
6H, C6H6), 2.91 (s, 3H, NMe), 2.71 (s, 3H, NMe), 2.72 (d, 1H,
2JHH ¼ 14 Hz, H5), 2.39 (d, 1H, 2JHH ¼ 14, Hz, 2H, H6). 31P{1H} NMR
2.4.2.1. Yield. 368 mg, 70%. 1H NMR (CD3COCD3): 9.23 (d,
3JHH ¼ 5.8 Hz, 1H, H8), 8.18 (d, 1H, 3JHH ¼ 8.1 Hz, H1), 8.06e7.91 (m,
3H, H2, H3 and H6), 7.26 (t, 1H, 3JHH ¼ 7.3 Hz, H4), 7.21e7.12 (m, 2H,
H5 and H7), 6.16 (s, 6H, C6H6), 1.53e1.11 (m, 18H, PBu3), 0.8 (t,
3J ¼ 7.0 Hz, 9H, PBu3). 31P{1H} NMR (CD3COCD3): 21.39 (s,
PBu3): ꢃ144 (stp, J ¼ 706.6 PF6). M/S FABþ: 536 ([M þ H] e PF6]:
(100%), 458 [(M þ H) e (C6H6 þ PF6)] (28%), 334 [(M þ H) e
(PBu3 þ PF6)] (18%), 256 [(M þ H) e (C6H6 þ PBu3 þ PF6)] (19%). IR
1
(CD3CN): 33.5 (s, PPh3), ꢃ144.6 (stp, JPF ¼ 704.25 Hz, PF6). M/S
FABþ: 576 [(M þ H) e PF6] (18%), 498 [(M þ H) e (C6H6 þ PF6)]
(12%), 314 [(M þ H) e (PPh3 þ PF6)] (15%), 236 [(M þ H) e
(C6H6 þ PPh3 þ PF6)] (6%). IR (FTIR):
y 832 (PF6, s). Anal. Calcd. for
(FTIR):
y 832 (PF6, s). Anal. Calcd. for C29H41F6NP2Ru: C, 51.17; H,
C
33H33F6NP2Ru: C, 55.00; H, 4.62; N, 1.94. Found: C, 54.72; H, 4.65;
6.07; N, 2.06. Found: C, 51.26; H, 5.91; N, 2.08.
N, 2.12.
2.4.3. Synthesis of 2c
2.3.2. Synthesis of 1b
2.3.2.1. Yield. 184 mg, 70%. 1H NMR (CD3CN): 7.79 (d, 1H,
3JHH ¼ 9 Hz, H1), 7.11e6.85 (m, 3H, H2þH3þH4), 5.76 (s, 6H, C6H6),
3.59 (d, 1H, 2JHH ¼ 12 Hz, H5), 3.31 (d, 1H, 2JHH ¼ 12 Hz, H6), 2.93 (s,
3H, NMe), 2.74 (s, 3H, NMe), 1.84e1.53 (m, 6H, PBu3), 1.43e1.20 (m,
12H, PBu3), 0.90 (t, 3JHH ¼ 6.8 Hz, 9H, PBu3). 31P{1H} NMR (CD3CN):
2.4.3.1. Yield. 397 mg, 88%. 1H NMR (CD3CN) 9.0 (d, 1H, 3JHH ¼ 6 Hz,
3
H8), 7.83 (d, 1H, JHH ¼ Hz, H1), 7.56e7.35 (m, 6H, H2 to H7),
7.30e7.25 (m, 6H, SbPh3), 7.05e6.9 (m, 9H, SbPh3), 5.90 (s, 6H,
C6H6). 31P{1H} NMR (CD3CN): ꢃ143.99 (stp, 1JPF ¼ 704.25 PF6). M/S
FABþ: 687 [(M þ H) e PF6] (35%), 609 [(M þ H) e (C6H6 e PF6)] (1%),
13.02 (s, PBu3), ꢃ144 (stp, JPF ¼ 704.6 Hz, PF6). M/S FABþ: 516
1
334 [(M
þ
H)
e
(SbPh3
e
PF6)] (92%), 256 [(M
830 (PF6, s). Anal. Calcd. for
35H29F6NPRuSb$0.5CH2Cl2: C, 48.79; H, 3.46; N, 1.60. Found: C,
þ
H) e
[(M þ H) e PF6] (100%), 438 [(M þ H) e (C6H6 þ PF6)] (34%), 314
(C6H6 þ SbPh3 þ PF6)] (25%). IR (FTIR):
y
[(M
þ
H)
e
(PBu3
þ
PF6)] (96%), 236 [(M
þ
H) e
C
(C6H6 þ PBu3 þ PF6)] (30%). IR (FTIR):
y 831 (PF6, s). Anal. Calcd. for
48.77; H, 3.47; N, 1.67.
C
27H45F6NP2Ru$0.5CH2Cl2: C, 46.98; H, 6.59; N, 1.99. Found: C,
47.41; H, 6.51; N, 2.21.
2.5. Synthesis of 2d
2.3.3. Synthesis of 1c
A solution of complex 3′ (200 mg, 0.543 mmol) and AgPF6
(137.4 mg, 0.543 mmol) in 25 mL of acetonitrile was stirred for 2 h
at room temperature. The reaction mixture was filtered through
Celite and the remaining pale yellow solution was evaporated to
dryness. Crystallization from CH2Cl2/diethyl ether (slow diffusion)
gave yellow crystals, which were washed with diethyl ether and
dried under vacuum. Yield: 197 mg, 70%. 1H NMR (CD3CN): 9.24
2.3.3.1. Yield. 275 mg, 85%. 1H NMR (CD3CN): 8.05 (d, 1H,
3JHH ¼ 6 Hz, H1), 7.53 (t, 3H, JHH ¼ 7.5 Hz, H9), 7.44 (t, 6H,
3
3JHH ¼ 7 Hz, H9), 7.32 (d, 6H, JHH ¼ 7.5 Hz, H9), 7.16 (t, 1H,
3
3JHH ¼ 7.5 Hz, H2), 6.96 (t, 1H, JHH ¼ 7.5 Hz, H3), 6.69 (d, 1H,
3
3JHH ¼ 7 Hz, H4), 5.79 (s, 6H, C6H6), 3.17 (d, 1H, JHH ¼ 15 Hz, H5),
2
3.04 (s, 3H, NMe), 2.93 (d, 1H, 3JHH ¼ 12 Hz, H6), 2.74 (s, 3H, NMe).
31P{1H} NMR (CD3CN): ꢃ144 (stp, 1JPF ¼ 706.5 Hz, PF6). M/S FABþ:
667 [(M þ H) e PF6] (12%), 589 [(M þ H) e (C6H6 þ PF6)] (2%), 314
3
4
3
(dd, 1H, JHH ¼ 6 Hz, JHH ¼ 2 Hz, H8), 8.17 (ddd, 1H, JHH ¼ 8 Hz,
5
4JHH ¼ 1.5, JHH ¼ 0.6 Hz, H1), 7.94e7.90 (m, 2H, H3 and H5), 7.77
(dd,1H, 3JHH ¼ 8 Hz, 4JHH ¼ 1.5 Hz, H7), 7.28e7.15 (m, 3H, H2, H4 and
H6), 5.78 (s, 6H, C6H6), 2.16 (s, 3H, NCMe). 31P{1H} NMR
(CD3CN): ꢃ144 (stp, 1JPF ¼ 704.25 Hz, PF6). M/S FABþ: 375[(M þ H)
e PF6] (1%), 297 [(M þ H) e (C6H6 þ PF6)] (2%), 334 [(M þ H) e
(NCMe þ PF6)] (32%), 256 [(M þ H) e (C6H6 þ NCMe þ PF6)] (8%). IR
[(M
(C6H6 þ SbPh3 þ PF6)] (10%). IR (FTIR):
33H33F6NPRuSb: C, 48.85; H, 4.10; N, 1.73; Found: C, 48.83; H, 4.10;
N, 1.87.
þ
H)
e
(SbPh3
þ
PF6)] (45%), 236 [(M
þ
H) e
y
831 (PF6, s). Anal.: Calcd. for
C
2.4. General procedure for complexes 2aec
(FTIR):
y 827 (PF6, s), 2261 (NCMe, w). Anal. Calcd. for
C19H17F6N2PRu: C, 43.94; H, 3.30; N, 5.39. Found: C, 43.83; H, 3.22;
N, 5.30.
In a typical experiment, a solution of 3’ (200 mg, 0.543 mmol),
0.815 mmol of ligand (2a triphenylphosphine, 2b tri-n-butylphos-
phine or 2c triphenylstibine) and NH4PF6 (177 mg, 1.086 mmol) in
30 mL of MeOH was stirred for 20 h at 45 ꢀC. The solvent was
evaporated under vacuum, and the residue was dissolved in CH2Cl2.
1
Although these results are outside the range viewed as establishing analytical
purity, they are provided to illustrate the best values obtained to date.

V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
303
2.6. Procedures for the polymerization reactions
product) and [Ru(Phpy)(NCMe)4]PF6, where the benzene ring has
been substituted by three acetonitrile molecules [18]. In order to
specifically prepare 2d, a different strategy was chosen and a
2.6.1. Polymerization of styrene
All reactions were carried out in solution (St/MEK 50% v/v) un-
der dinitrogen or argon atmosphere at 80 ꢀC. The initial molar ratio
of [monomer]0/[initiator]0/[complex]0 was 200/1/1 with BEB used
as initiator. A polymerization procedure was as follows: 3′ (50 mg,
chloride abstraction from 3′ (prepared from [Ru(h
6-C6H6)Cl2]2 and
2-phenylpyridine in MeOH [40]) by silver hexafluorophosphate in
acetonitrile led to pure 2d in 70% yield. To synthesize the remaining
members of the 2 series, the chloride in 3′ was easily substituted by
the corresponding aec ligands in the presence of NH4PF6 in MeOH
and complexes 2aec were obtained in high yields (70e88%). It is
worth noting that compounds 1b, 2a and 2c crystallized with some
0.135 mmol), BEB (18
(50 mg, 0.13 mmol), BEB (18
m
L, 0.135 mmol), St (3 mL, 27 mmol) or 300
L, 0.13 mmol), St (3 mL, 27 mmol);
m
MEK (3 mL) and n-decane (0.3 mL) were placed in a 25 mL Schlenk
flask and degassed three times using pump-nitrogen cycles. The
mixture was stirred for 10 min at room temperature until ho-
mogenous. The flask was immersed in an oil bath previously heated
at 80 ꢀC. Samples were removed after certain time intervals using
degassed syringes. For the cationic complexes the procedure was as
amount of dichloromethane as detected by 1H NMR at
d
¼ 5.48
(CD3CN), 5.64 (acetone-d6) and 5.44 (CD3CN) respectively. Those
chemicals shifts are in accordance with the reported values [46]. As
such, the presence of dichloromethane is reflected in the obtained
elemental analysis values.
follows: 2a (100 mg, 0.135 mmol), BEB (18 mL, 0.135 mmol), St (3 mL,
27 mmol), n-decane (0.3 mL) and MEK (3 mL) were placed in a
25 mL Schlenk flask and degassed three times using pump-nitrogen
cycles. The mixture was stirred for 10 min at room temperature
until homogenous, and 8 aliquots (0.75 mL each) were injected into
baked glass tubes and sealed under nitrogen. The tubes were
immersed in an oil bath previously heated at 80 ꢀC. The polymer-
izations were stopped at the desired time by cooling the tubes in
ice-cold water. Conversions were determined by GC and the poly-
mer samples injected into GPC without purification.
3.2. X-ray diffraction studies
Single crystals suitable for X-ray diffraction were obtained for all
new complexes by slow diffusion of diethylether into a dichloro-
methane solution. Compound 1b crystallized with one molecule of
dichloromethane. Crystallographic data, relevant bond distances
and angles are summarized in Tables 1e4. The molecular structures
are depicted in Figs. 2 and 3. Structure of complex 1d has previously
been reported [19], however, for the sake of comparison, crystal-
lographic data have been included in Tables 3 and 4 All compounds
adopt the expected “three-legged piano stool” structure. The dis-
tances between the nitrogen of the metalated ligand and the
ruthenium center are slightly longer for the dmba derivatives than
for their phpy counterparts (from 0.13 Å difference between 1a and
2a to 0.007 Å between 1c and 2c). This can be explained by the
nature of the nitrogen bound to the metal, as due to the back-
bonding effects, pyridines are usually more strongly bound to RuII
than tertiary amines. As expected the antimony-ruthenium bonds
are about 0.25 Å longer than the phosphorous-ruthenium bonds,
reflecting the increase in the covalent radii from phosphorus
(1.10 Å) to antimony (1.41 Å), while almost no difference is observed
in the phosphorous-ruthenium bonds for complexes bearing PPh3
or PBu3 [45]. The average distance between the centroid of the
benzene ring for all compounds is 1.72 Å, and does not vary
significantly between the 1 and 2 series, albeit slightly shorter for
1d and 2d bearing acetonitrile ligand. These distances are consis-
2.6.2. Polymerization of vinyl acetate
The polymerization of VAc was performed as follows: complex
1b (178 mg, 0.271 mmol), CCl4 (26 mL, 0.271 mmol), VAc (5 mL,
54.2 mmol), n-decane (0.5 mL) and MEK (5 mL) were placed in a
25 mL Schlenk flask and degassed three times using pump-nitrogen
cycles. The mixture was stirred for 10 min at room temperature
until homogenous, and 8 aliquots (0.62 mL each) of solution were
injected into baked glass tubes and sealed under nitrogen. The
tubes were immersed in an oil bath previously heated at 80 ꢀC. The
polymerizations were stopped at the desired time by cooling the
tubes in ice-cold water. Conversions were determined by GC and
the polymer samples injected into GPC without purification.
3. Results and discussion
3.1. Synthesis
tent with those reported for similar (
plexes [47].
h
6-arene)ruthenium(II) com-
As mentioned above, the complexes used in the study could be
divided into three principal groups based on the cyclometalated
ligand, dmba (1) vs phpy (2 and 3), and on the charge, cationic (1
and 2) vs neutral (3) compounds. In addition, all complexes bear h6
coordinated benzene, and four different donor ligands were
introduced within the dmba and phpy series, as shown in Fig. 1,
from relatively labile MeCN to more strongly bound phosphines
and stibines [45]. The ligands were also chosen according to their
steric and electronic properties, for instance bulkier PPh3 and SbPh3
vs MeCN and P(n-Bu)3, or more labile stibine vs phosphine [45].
Additionally, two neutral ruthenacycles derived from phpy (3′) and
tolpy (300) were also used in the present study [40]. Complex 1d can
readily be prepared from the direct metalation of N,N-dime-
As for angles, no important difference between the two series
can be noted around the ruthenium center, except for the cyclo-
metalated nitrogen e ruthenium e E angles [E ¼ P (a, b), Sb (c), N
(d)], which are larger for dmba, from a 1.60ꢀ (d) difference to 10.05ꢀ
(b). Those data probably reflect some steric repulsion between the
phosphine or stibine ligands and the methyl groups of the dime-
thylamino substituent as the smallest difference is observed for
acetonitrile and the highest for bulkier P(n-Bu)3.
3.3. Polymerization studies
It should be noted that the complexes, particularly the cationic
ones, are poorly soluble in non-polar organic solvents, and MEK
was found to be the solvent of choice. The ruthenacycles are well
soluble in MEK, it is non-coordinating, it allows increasing the
temperature up to 80 ꢀC and can be readily evaporated making
working-up relatively straightforward. We started the investigation
with polymerization of styrene, using BEB as the initiator. All
complexes were evaluated in this polymerization for 6 and 24 h and
the best candidates for the detailed kinetic studies were selected
according to these preliminary data. The data are given in Table 5.
thylbenzylamine by the [Ru(h
6-C6H6)Cl2]2 dimer in acetonitrile, as
reported by Pfeffer et al. [18]. Compounds 1aec were obtained in
excellent yields (70e85%) from 1d by substitution in methanol of
the coordinated acetonitrile by more basic ligands such as triphe-
nylphosphine (a) tri-n-butylphosphine (b) and triphenylstibine (c)
[14].
Pfeffer also reported that the reaction between 2-
phenylpyridine and the ruthenium dimer under the same condi-
tions does not produce a clean reaction but a mixture of 2d (minor

304
V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
Table 1
Crystal structure data for complexes 1aec.
1a
1b
1c
Empirical formula
Formula weight
Temperature (K)
Wavelength (Å)
Crystal system
Space group
C33H33F6NP2Ru
720.61
298(2)
0.71073
monoclinic
P21/c
C55H92Cl2F12N2P4Ru2
1406.23
298(2)
0.71073
triclinic
C33H33F6NPRuSb
811.39
298(2)
0.71073
monoclinic
P21/c
P-1
Unit cell dimensions (in Å and ꢀ
)
a ¼ 11.5652(10)
b ¼ 17.5904(14)
c ¼ 15.4768(13)
a ¼ 10.2217(9)
b ¼ 12.6684(11)
c ¼ 13.2789(12)
a ¼ 17.650(5)
b ¼ 10.360(3)
c ¼ 18.661(5)
a
b
g
¼ 90
a
b
g
¼ 100.2230(10
¼ 100.5090(10)
¼ 97.099(2)
a
b
g
¼ 90
¼ 105.7810(10)
¼ 90
¼ 112.841(4)
¼ 90
Volume (Å3)
3029.9(4)
1642.1(3)
3144.8(15)
Z
4
1
4
Density (mg/m3, Calcd.)
1.580
0.685
1464
1.422
0.708
726
1.714
1.450
1608
Absorption coeff. (mmꢃ1
F(000)
)
Crystal size (mm)
q
Index ranges
0.32 ꢁ 0.20 ꢁ 0.15
0.38 ꢁ 0.12 ꢁ 0.1
1.66 to 25.38
ꢃ12 ꢂ h ꢂ 12
ꢃ15 ꢂ k ꢂ 15
ꢃ16 ꢂ l ꢂ 16
13,685
0.34 ꢁ 0.09 ꢁ 0.05
2.21 to 25.41
ꢃ21 ꢂ h ꢂ 21
ꢃ12 ꢂ k ꢂ 12
ꢃ22 ꢂ l ꢂ 22
25,448
range for data collection (ꢀ)
1.79e25.35ꢀ
ꢃ13 ꢂ h ꢂ 13
ꢃ21 ꢂ k ꢂ 21
ꢃ18 ꢂ l ꢂ 18
24,735
Reflections collected
Independent reflections
Absorption correction
Refinement method
5529 [R(int) ¼ 0.0576]
6006 [R(int) ¼ 0.0324]
5785 [R(int) ¼ 0.0797]
None
analytical
analytical
full-matrix least-squares on F2
5529/417/445
0.910
full-matrix least-squares on F2
6006/530/460
0.955
full-matrix least-squares on F2
5785/526/471
0.915
Data/Restrains/Parameters
Goodness of fit on F2
Final R indices [I > 2
R indices (all data)
Largest diff. peak and hole (e. Å)
s
(I)]
R1 ¼ 0.0384, wR2 ¼ 0.0721
R1 ¼ 0.0574, wR2 ¼ 0.0775
0.651 and ꢃ0.372
R1 ¼ 0.0381, wR2 ¼ 0.0846
R1 ¼ 0.0489, wR2 ¼ 0.0885
0.491 and ꢃ0.414
R1 ¼ 0.0516, wR2 ¼ 0.0991
R1 ¼ 0.0849, wR2 ¼ 0.1090
3.359 and ꢃ1.054
Table 2
Crystal structure data for complexes 2aed.
2a
2b
2c
2d
Empirical formula
Formula weight
C35H29F6NP2Ru
740.60
C29H41F6P2Ru
680.64
C35H29F6NPRuSb
831.38
C19H17F6N2PRu
519.39
Temperature (K)
Wavelength (Å)
298(2)
0.71073
298(2)
0.71073
298(2)
0.71073
298(2) K
0.71073
Crystal system
Space group
Unit cell dimensions (in Å and ꢀ
monoclinic
P21/c
monoclinic
P21/n
monoclinic
P21/n
Orthorhombic
P b c n
)
a ¼ 11.4000(9)
b ¼ 17.2050(13)
c ¼ 15.9667(12)
a ¼ 8.7804(13)
b ¼ 16.256(2)
c ¼ 21.947(3)
a ¼ 10.2781(12)
b ¼ 12.0941(14)
c ¼ 26.303(3)
a ¼ 15.9934(16)
b ¼ 16.3803(17)
c ¼ 15.8062(16)
a
b
g
¼ 90
a
b
g
¼ 90
a
b
g
¼ 90
a
b
g
¼ 90
¼ 90
¼ 90
¼ 97.7140(10)
¼ 90
¼ 94.240(2)
¼ 90
¼ 93.359(2)
¼ 90
Volume (Å3)
3103.3(4)
4
3123.9(8)
4
3264.0(7)
4140.9(7)
Z
4
8
Density (mg/m3, Calcd.)
1.585
0.671
1496
1.447
0.659
1400
1.692
1.399
1640
1.666
0.894
2064
Absorption coeff. (mmꢃ1
F(000)
)
Crystal size (mm)
q
Index ranges
0.30 ꢁ 0.20 ꢁ 0.10
1.75 to 25.35
ꢃ13 ꢂ h ꢂ 13
ꢃ20 ꢂ k ꢂ 20
ꢃ19 ꢂ l ꢂ 19
25,420
0.36 ꢁ 0.1 ꢁ 0.08
1.86 to 25.35
ꢃ10 ꢂ h ꢂ 10
ꢃ19 ꢂ k ꢂ 19
ꢃ26 ꢂ l ꢂ 26
25,302
0.31 ꢁ 0.14 ꢁ 0.05
1.85 to 25.35
ꢃ12 ꢂ h ꢂ 12
ꢃ14 ꢂ k ꢂ 14
ꢃ31 ꢂ l ꢂ 31
35,321
0.34 ꢁ 0.26 ꢁ 0.20
1.78 to 25.40
ꢃ19 ꢂ h ꢂ 19
ꢃ19 ꢂ k ꢂ 19
ꢃ19 ꢂ l ꢂ 19
32,295
range for data collection (ꢀ)
Reflections collected
Independent reflections
5678
5696
5973 [R(int) ¼ 0.0940]
3804 [R(int) ¼ 0.0583]
[R(int) ¼ 0.0645]
[R(int) ¼ 0.0898]
Absorption correction
Refinement method
None
none
analytical
analytical
full-matrix least-squares on F2 full-matrix least-squares on F2 full-matrix least-squares on F2 full-matrix least-squares on F2
Data/Restrains/Parameters
5678/399/461
0.882
5696/675/474
0.870
5973/423/461
0.817
3804/1226/483
1.046
Goodness of fit on F2
Final R indices [I > 2
R indices (all data)
Largest diff. peak and hole (e. Å)
s
(I)]
R1 ¼ 0.0381, wR2 ¼ 0.0673
R1 ¼ 0.0617, wR2 ¼ 0.0730
0.573 and ꢃ0.330
R1 ¼ 0.0483, wR2 ¼ 0.0752
R1 ¼ 0.0958, wR2 ¼ 0.0854
0.662 and ꢃ0.474
R1 ¼ 0.0427, wR2 ¼ 0.0637
R1 ¼ 0.0882, wR2 ¼ 0.0723
0.727 and ꢃ0.371
R1 ¼ 0.0314, wR2 ¼ 0.0773
R1 ¼ 0.0488, wR2 ¼ 0.0857
0.231 and ꢃ0.433

V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
305
Table 3
Comparison of selected bond distances [Å] between 1aed and 2aed complexes.
1a
2a
1b
2b
1c
2c
1d [19]
2d
a
RueC6H6
RueCmet
RueNmet
RueE
1.782
1.755
1.758
1.755
1.742
1.742
1.711
1.709
2.069(3)
2.206(3)
2.3603(9)
2.058(3)
2.076(3)
2.3324(9)
2.067(3)
2.200(3)
2.3534(9)
2.038(5)
2.074(3)
2.339(1)
2.077(7)
2.094(7)
2.5986(9)
2.063(6)
2.087(5)
2.5815(7)
2.069(2)
2.170(2)
2.058(2)
2.062(3)
2.081(2)
2.046(2)
E ¼ P (a, b), Sb (c), N (d).
a
Centroid.
Table 4
Selected angles [ꢀ] for the complexes.
1a
2a
1b
2b
1c
2c
1d [19]
2d
NeRueC
NeRueE
78.2(1)
95.94(7)
85.97(8)
126.83
77.85(13)
88.57(7)
84.41(9)
132.05
78.30(12)
98.85(7)
86.35(8)
123.73
78.28(18)
88.8(1)
85.6(1)
130.66
78.8(3)
91.8(3)
82.05(17)
126.75
77.6(2)
85.96(1)
82.26(13)
131.82
78.10(10)
86.65(9)
85.98(9)
126.83
78.27(10)
85.05(9)
84.54(10)
129.60
CeRueE
C6H6aeRueE
E ¼ P (a, b), Sb (c), N (d).
a
Centroid.
Fig. 2. ORTEP views of complexes 1aec. Thermal ellipsoids are drawn with 40% probability level. Hydrogen atoms and PF-6 anions are omitted for clarity.
The highest conversions, above 60% in 24 h, were obtained with
neutral compounds, 3′ and 300. The tolpy (300) derivative was slightly
more active but the difference in the final yields was not very sig-
nificant. However, no dependence of the molecular weight on
conversion was observed. Polymers of ca. 25,000 Da were obtained
within 6 h of the reaction and this value did not change after 24 h,
even if the conversion significantly increased within this period of
time. Conversions reached in the polymerizations mediated by
other complexes were lower, usually around 50% after 24 h with a
few exceptions as described below. Thus, complexes bearing dmba
(1aed) showed quite a similar activity, resulting in around 50%
polymer yields in 24 h and producing high disperse (around 2)
polySt with molecular weights varying from ca. 15,000 to
28,000 Da. The growth of the molecular weights with conversion
was observed for the polymerizations mediated by 1b and 1d, even
if the dispersity indexes were high and the molecular weights
determined by GPC were higher than the calculated values. In
general, the complexes were quite active taking into account the
relatively low temperature for the polymerization of St [48]. The
largest difference in the catalytic performance was noted within
compounds of group 2, phpy-based ruthenacycles. The lowest
polymer yields (16%) were obtained using complexes with SbPh3
and relatively more labile MeCN ligands and those polymers were
also characterized by the highest molecular weights (40,000 and
60,000 Da respectively). Complexes bearing phosphine ligands
resulted in higher conversions after 24 h, but the polySt obtained
with complex 2a (PPh3) displayed a very low molecular weight and
high dispersity index. In fact, the most promising results were

306
V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
Fig. 3. ORTEP views of complexes 2aed. Thermal ellipsoids are draw with 40% probability level. Hydrogen atoms and PFꢃ6 anions are omitted for clarity.
Table 5
Results of the polymerization of styrene in the presence of the cyclometalated ruthenium compounds.
Complex
Time (hours)
Conversion (%)
M
n(GPC)*10ꢃ3 (dalton)
Mn(Theo)*10ꢃ3 (dalton)
Mw/Mn
1a
6
24
6
24
6
24
6
24
6
24
6
24
6
24
6
24
6
43.5
55.0
37.0
50.0
35.0
50.5
40.4
52.5
5.0
20.5
32.0
46.0
2.0
16.4
12.0
16.5
28.7
62.9
48.6
68.2
15.6
16.6
19.0
28.3
16.5
17.4
14.6
18.4
e
9.1
11.5
7.7
10.4
7.2
10.5
8.4
10.9
e
1.70
1.97
1.90
2.36
1.74
1.90
1.80
1.72
e
1b
1c
1d
2a
2b
2c
2d
3′
2.2
8.4
10.5
e
4.3
6.7
9.5
e
2.74
1.40
1.52
e
43.5
33.0
50.6
23.4
25.9
24.5
26.1
3.3
2.5
3.4
5.9
13.2
10.2
14.2
2.38
1.84
2.47
1.89
1.99
2.22
2.31
24
6
24
300
obtained using complex with P(n-Bu)3 ligand 2b. The molecular
weights of polySt formed in the reaction mediated by this ruth-
enacycle coincided well with the calculated molecular weights, and
the Mw/Mn values were also the lowest in comparison with those
obtained with the other complexes.
It is important to add that the radical character of the reaction
was established by the stable radical (TEMPO) methodology. When
5 equivalents of TEMPO were added into the reaction mixture, no
polymer was obtained after 24 h of heating. Besides, no polymer-
izations could be observed without the initiator, even in the
presence of the ruthenacycles.
Thus taking into account these preliminary data the following
complexes were selected for further investigation: cationic 1b and
1d from the dmba-based group as they were the only catalysts
showing the growth of the molecular weights with conversion,
their analogs 2b and 2d from phpy-based ruthenacycles, and both
neutral complexes (3′ and 300) as they showed the highest
conversions.
The results of the kinetic measurements of polymerization with
these compounds are presented in Table 6. The molecular weights

V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
307
Table 6
Data on the polymerization of styrene mediated by selected complexes.
Complex
Time (hours)
Loss of initiator (%)
Conversion (%)
M
n(GPC)*10ꢃ3 (dalton)
Mn(Theo)*10ꢃ3 (dalton)
Mw/Mn
1
1b
2
1
3
6
4
e
5
e
6
e
7
e
2
3
6
12
1
18
98
e
2.9
e
e
e
24.0
37.0
48.8
e
15.3
19.0
27.0
e
5.0
7.7
10.2
e
1.79
2.08
2.09
e
e
1d
2b
2d
3′
2
2
3
6
12
1
2
3
6
12
1
2
3
6
12
1
2
3
6
12
1
16
95
e
e
e
e
e
31.8
40.4
51.2
e
8.0
14.6
17.8
e
6.6
8.4
10.6
e
1.91
1.80
1.64
e
e
4
14
99
e
2.0
e
e
e
18.5
32.5
45.0
e
5.3
8.4
10.5
e
3.8
6.7
9.4
e
2.00
1.42
1.39
e
e
7
19
94
e
e
e
e
e
9.3
27.7
33.0
50.8
e
1.9
2.5
3.3
e
2.14
2.32
2.43
e
12.0
16.0
4.5
e
15
27
50
60
75
15
30
54
62
80
8.0
e
e
e
15.2
28.5
46.5
7.0
15.5
25.5
46.0
57.9
22.8
23.4
23.0
3.2
5.9
9.6
1.96
1.89
1.99
300
2
3
6
12
22.3
24.5
24.2
5.3
9.5
12.0
1.95
2.39
2.37
grew with conversion in the polymerizations mediated by the ionic
ruthenacycles, but only in the case of 1d and 2b a good agreement
between GPC and calculated molecular weight values was
observed.
before proceeding, meanwhile the polymerization in the presence
of neutral 3′ and 300 took place smoothly from the beginning.
Interestingly, the induction time coincided very well with the
consumption of the initiator in the reaction (columns 3 and 4,
Table 6). During the first two hours of heating, initiator was barely
consumed and the polymerizations did not occur. Abrupt disap-
pearance of the initiator during the third hour was accompanied by
initiation of the polymerization. In contrast, in polymerizations
The polySt synthesized with this complex 2b was also distin-
guished by narrow dispersity indexes. All these results indicate that
2b displayed a better control of the polymerization than any other
ruthenacycles involved in the study. Kinetic plot in semilogarithmic
coordinates and evolution of the molecular weights with the con-
version for this complex are given in Fig. 4. As one can see, the
semilogarithmic plot is quite linear and the GPC curves are unim-
odal, relatively narrow and shifted to the higher molecular weights
as the polymerization progressed. Compound 1d also demon-
strated a certain level of control in the polymerization but the
difference between experimental and theoretical molecular
weights was larger and the polySt was more disperse than with 2b.
Although an increase in the molecular weights of the polystyrene
was observed in the process mediated by 1b, the experimental
values were about 3 times higher than the calculated ones. Con-
versions achieved in 12 h were similar to the conversions obtained
after 24 h, meaning that the polymerizations reached a plateau in
12 h and did not progress anymore. Yields of around 50% were
obtained for the polymerizations with ionic complexes, except for
the process mediated by 2d, where the conversion was very low.
Polymerizations were slightly faster for the dmba based ruthena-
cycles than for 2b, although the control was much better with the
latter compound.
mediated by neutral group
3 compounds, the initiator was
consumed gradually during the reaction time. It is logical to sup-
pose that the cationic complexes underwent some kind of rear-
rangement during the induction time, converting them into active
catalysts.
Loss of initiator in the polymerizations with other ionic ruth-
enacycles was also verified by GC. The only other complex showing
a similar 3 h consumption period was 1c with SbPh3 ligand. Three
other ruthenacycles behaved differently. The loss of initiator was
very slow in the reaction with phpy derivatives bearing PPh3 and
SbPh3 ligands, 2a and 2c, as practically no initiator was consumed
in the first 3 h, and less than 15% after 6 h. This is in good agreement
with the data on the reactions catalyzed by these complexes as no
polymer was detected after 6 h. On the contrary, the initiator
consumption was much faster for dmba-based ruthenacycle with
PPh3 1a, as almost 70% reacted during the first hour of the poly-
merization. However, this fast consumption was not reflected in a
better controlled polymerization.
The complexes 1b and 2b were tested in the polymerization of
VAc using the same conditions as those for styrene polymerization.
In contrast to styrene, the radical derived from VAc is very active
and thus forms strong bonds with the trapping end-group [49].
Despite all the progress in ATRP, no efficient catalytic system for
controlled polymerization of VAc has been reported, as the vast
majority of the transition metal catalysts are too “mild” to activate
Polymerizations conducted in the presence of neutral ruthena-
cycles were fast but proceeded without any control, resulting in a
polymer of constant molecular weight. From data depicted in
Table 6, another important difference between ionic and neutral
complexes could be noted. The polymerizations catalyzed by the
ionic ruthenacycles clearly marked an induction period of 3 h

308
V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
Fig. 4. (Left) Kinetic plot of polymerization of St mediated by 2b at 80 ꢀC in MEK (50/50 v/v) with the molar ratios of [St]0/[2b]/[BEB]0 ¼ 200/1/1; (Right) GPC traces of polySt
obtained.
strongly bound C-Halogen terminals. Only a few studies on the
possibility of such process have been described and one of the
recent reports was based on complex 1d which performed ATRP of
VAc with moderate level of control in DMSO using CCl4 as initiator
[37,50]. Although the polymerization did not proceed to high
conversions and the Mw/Mn values were not low enough, the
participation of the ATRP mechanism in the process could be
demonstrated. In the present study, an attempt has been made to
expand the number of potential candidates for such important
catalytic process. Nevertheless, none of complexes 1b and 2b were
active in BEB initiated polymerization of VAc in MEK at 80 ꢀC, i.e.
under the conditions used for styrene. However, when CCl4 was
utilized instead of BEB, polymerization of VAc was observed with
complex 1b. The reaction resulted in a maximum of ca. 30% con-
version in 12 h and polyVAc of monomodal weight distribution
with Mn,GPC ¼ 8500 Da and Mw/Mn ¼ 1.45. Interestingly, the con-
stant growth of the molecular weights was observed, as well as the
decrease of Mw/Mn values with the conversion. Since CCl4 was also
consumed slowly (50% in 3 h and 80% in 6 h), the reaction most
probably proceeded via a red-ox initiation catalyzed by 1d, with
CCl4 acting as initiator and chain transfer agent [37]. However,
participation of the ATRP mechanism could not be completely
excluded in this case. For comparison, complex 1d under these
conditions resulted in a maximum of 15% conversion, producing
low molecular weight (5000 Da) polyVAc. Thus, complex 1b may be
more promising for the living/controlled polymerization of VAc
than 1d, but thorough investigation of the mechanism and search
for optimal conditions are required.
3.4. Mechanistic studies
As no clear conclusion could be drawn from the results
described above and in order to understand what kind of reaction
may occur during the initial period of the polymerization of sty-
rene, the stoichiometric reaction between BEB (initiator) and all the
ionic ruthenium complexes was investigated at 80 ꢀC by 1H NMR
spectroscopy and practically no changes were noted after 3 h,
except for complex 1a. Therefore, the reaction of 1a with BEB was
studied in more details and its behavior was compared with that of
2a.
An interesting reaction took place when complex 1a was heated
to 80 ꢀC in MEK for 3 h. A red precipitate was observed, which was
filtered off (45% yield) and analyzed by 1H and 31P NMR. Spectra
showed that the signals corresponding to the dmba moiety had
disappeared and a new signal appeared in 31P NMR at 23.6 ppm.
Comparison with the data obtained from a pure sample prepared
accordingly to the literature [41,42] showed that the coordination
of two bromides from the initiator took place, leading to the for-
mation of neutral [Ru(h
6-C6H6)Br2(PPh3)] complex 4 (Fig. 5).
Analysis by GC/MS of the remaining solution after filtration of 4
only showed the presence of N,N-dimethylbenzylamine, residual
BEB and 1-phenylethanol, probably arising from the hydrolysis of
BEB. No coupling products from BEB and dmba could be detected.
Under the same condition, complex 2a did not react with BEB. We
previously reported the relative fragility of dmba complexes versus
their phpy analogs and the cleavage of the RueC
s bond of 1d by
MeOH with the concomitant arene substitution by 2,20-bipyridine
Fig. 5. Reaction between complex 1a and the BEB initiator.

V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
309
[8]. When compound 4 was used in the polymerization of styrene
under the same conditions as described above, no polymer could be
detected after 24 h, allowing to discard the hypothesis that 4 is the
active catalyst formed during the induction period.
and scholarship 6758 to V. Martinez Cornejo) is gratefully
acknowledged. The authors thank Gerardo Cedillo Valverde (Insti-
tuto de Investigaciones en Materiales) and María de la Paz Orta
ꢀ
Perez (Instituto de Química) for the support in NMR and elemental
In another series of reactions, complex 2a was reacted with five
equivalents of styrene in (CD3)2CO at 80 ꢀC in a sealed tube, and the
reaction was monitored by 1H and 31P NMR. After 24 h, most of the
complex remained intact (approx. 80%), but some new peaks arose.
Significantly, in 31P NMR a new signal appeared at 22.65 ppm and in
1H NMR a signal at 7.34 ppm could be attributed to free benzene,
and new features were detected at 6.48 (dd), 6.42 (dd) and 5.23 (d)
ppm. Other new peaks are masked by those of 2a and styrene. This
observation suggests that styrene could be coordinated to the
ruthenium center, as the observed chemical shifts are in accordance
analysis measurements. We wish to dedicate this contribution to
the memory of Professor Armando Cabrera Ortiz (1944e2014).
Appendix A. Supplementary material
CCDC 1420624 (1a), 1420625 (1b), 1420626 (1c), 1420627 (2a),
1420628 (2b), 1420629 (2c) and 1420630 (2d) contain the sup-
plementary crystallographic data for the new compounds. These
data can be obtained free of charge from the Cambridge Crystal-
lographic data Center via www.ccdc.cam.ac.uk/data_request/cif.
with the data reported for (h
6-styrene)ruthenium complexes
[51,52]. The low yield after 24 h can be explained by the 1:5 (Ru:St)
stoichiometry of the reaction. Under the polymerization conditions
the Ru:St relation is 1:200, which could lead to a faster and more
quantitative reaction, as arene exchange on ruthenium(II) com-
plexes usually takes place with a large excess of the incoming
ligand at high temperatures [53]. Thus, an arene exchange reaction
could explain the low catalytic activity displayed by the complexes.
However, those are preliminary results and more work is needed in
order to isolate the new complex and confirm the coordination of
styrene.
References
[1] J.-P. Djukic, J.-B. Sortais, L. Barloy, M. Pfeffer, Eur. J. Inorg. Chem. (2009)
817e853.
[2] M. Albrecht, Chem. Rev. 110 (2010) 576e623.
[3] A.D. Ryabov, Chem. Rev. 90 (1990) 403e424.
[4] Palladacycles: Synthesis, Characterization and Applications, John Wiley &
Sons, 2008.
[5] S.H. Wadman, J.M. Kroon, K. Bakker, M. Lutz, A.L. Spek, G.P.M. van Klink, G. van
Koten, Chem. Commun. 4 (2007) 1907.
[6] T. Bessho, E. Yoneda, J.-H. Yum, M. Guglielmi, I. Tavernelli, H. Imai,
€
U. Rothlisberger, M.K. Nazeeruddin, M. Gratzel, J. Am. Chem. Soc. 131 (2009)
5930e5934.
[7] J.-P. Launay, Chem. Soc. Rev. 30 (2001) 386e397.
[8] A.D. Ryabov, V.S. Sukharev, L. Alexandrova, R. Le Lagadec, M. Pfeffer, Inorg.
Chem. 40 (2001) 6529e6532.
4. Conclusions
[9] A.D. Ryabov, V.S. Kurova, E.V. Ivanova, R. Le Lagadec, L. Alexandrova, Anal.
Chem. 77 (2005) 1132e1139.
A series of [Ru(h
6-C6H6)(N∩C)L]PF6 isomorphous half-sandwich
[10] E.V. Ivanova, I.V. Kurnikov, A. Fischer, L. Alexandrova, A.D. Ryabov, J. Mol.
Catal. B Enzym 41 (2006) 110e116.
complexes, bearing cyclometalated dmba (1aed) and phpy (2aed)
moieties, has readily been prepared by ligand substitution in high
yields. The comparative catalytic activity of the complexes in the
radical polymerization of styrene and vinyl acetate has been stud-
ied. However, no direct correlation between the catalyst structure
and its catalytic activity could be drawn. The complexes of the
dmba group showed good activity in the polymerization of styrene
but only with two of them, 1b (P(n-Bu)3) and 1d (MeCN), a growth
of the molecular weight with conversion was observed. Complexes
of the phpy group were much less active, except for 2b. Although
the polymerization of styrene with this complex was somewhat
slower, a good coincidence between the experimental and calcu-
lated molecular weights was found, meaning a better controlled
process. Kinetic studies revealed an induction period in the poly-
merization of styrene mediated by ionic complexes. The induction
period coincides with the complete consumption of the initiator,
suggesting some rearrangement in the coordination sphere of the
complexes converting them into active catalysts. Such gradual
rearrangement may be responsible for the high dispersity indexes
of the obtained polySt. As for the polymerization of vinyl acetate,
only complex 1b with P(n-Bu)3, was active, providing CCl4 is used as
initiator. Interestingly, the complexes exhibiting the best control/
activity behavior, 2b in the polymerization of St and 1b in the
polymerization of VAc, contain the P(n-Bu)3 ligand. According to
Tolman [54], P(n-Bu)3 is more electron donating and less bulky than
the much more commonly used PPh3, making worth investigating
the contribution of such parameters in the mechanism of ATRP and
related reactions catalyzed by ruthenium(II) species. On the other
hand, displacement of the metalated dmba fragment by bromides
from the initiator was observed, but the resulting neutral species
did not catalyze the polymerization, suggesting that more robust
phpy ligand may be preferable.
[11] A.D. Ryabov, Adv. Inorg. Chem. 55 (2004) 201e269.
[12] H. Huang, P. Zhang, H. Chen, L. Ji, H. Chao, Chem. e A Eur. J. 21 (2015)
715e725.
~
[13] B. Pena, A. David, C. Pavani, M.S. Baptista, J.P. Pellois, C. Turro, K.R. Dunbar,
Organometallics 33 (2014) 1100e1103.
[14] L. Leyva, C. Sirlin, L. Rubio, C. Franco, R. Le Lagadec, J. Spencer, P. Bischoff,
C. Gaiddon, J.P. Loeffler, M. Pfeffer, Eur. J. Inorg. Chem. (2007) 3055e3066.
[15] L.N. Lewis, J.F. Smith, J. Am. Chem. Soc. 108 (1986) 2728e2735.
[16] P. Dani, T. Karlen, R.A. Gossage, S. Gladiali, G. Van Koten, Angew. Chem. Int. Ed.
Engl. 39 (2000) 743e745.
[17] W. Baratta, P. Da Ros, A. Del Zotto, A. Sechi, E. Zangrando, P. Rigo, Angew.
Chem. Int. Ed. Engl. 43 (2004) 3584e3588.
[18] S. Fernandez, M. Pfeffer, V. Ritleng, C. Sirlin, Organometallics 18 (1999)
2390e2394.
ꢁ
[19] R. Le Lagadec, L. Rubio, L. Alexandrova, R.A. Toscano, E.V. Ivanova, R. Meskys,
ꢁ
V. Laurinavicius, M. Pfeffer, A.D. Ryabov, J. Organomet. Chem. 689 (2004)
4820e4832.
[20] B. Boff, M. Ali, L. Alexandrova, N.A. Espinosa-Jalapa, R.O. Saavedra-Díaz, R. Le
Lagadec, M. Pfeffer, Organometallics 32 (2013) 5092e5097.
ꢀ
[21] A.S. Lima-Barbosa, C. Werle, C.O. Colunga-Oliva, C. Franco-Rodríguez,
R.A. Toscano, R. Le Lagadec, M. Pfeffer, Inorg. Chem. 54 (2015) 7617e7626.
[22] M. Kato, M. Kamigaito, M. Sawamoto, T. Higashimuras, Macromolecules 28
(1995) 1721e1723.
[23] M. Ouchi, T. Terashima, M. Sawamoto, Chem. Rev. 109 (2009) 4963e5050.
[24] Y. Borguet, A. Richel, S. Delfosse, A. Leclerc, L. Delaude, A. Demonceau, Tet-
rahedron Lett. 48 (2007) 6334e6338.
[25] K. Severin, Chimia 66 (2012) 386e388.
[26] K. Parkhomenko, L. Barloy, J.B. Sortais, J.P. Djukic, M. Pfeffer, Tetrahedron Lett.
51 (2010) 822e825.
[27] W.A. Braunecker, N.V. Tsarevsky, A. Gennaro, K. Matyjaszewski, Macromole-
cules 42 (2009) 6348e6360.
[28] T. Ando, M. Kamigaito, M. Sawamoto, Macromolecules 33 (2000) 5825e5829.
[29] A. Richel, A. Demonceau, A.F. Noels, Tetrahedron Lett. 47 (2006) 2077e2081.
[30] C. Aguilar-Lugo, R. Le Lagadec, A.D. Ryabov, G.C. Valverde, S. Lopez-Morales,
L. Alexandrova, J. Polym. Sci. Part A Polym. Chem. 47 (2009) 3814e3828.
ꢀ
[31] M. Beley, S. Chodorowski-Kimmes, J.-P. Collin, P. Laine, J.-P. Launay, J.-
P. Sauvage, Angew. Chem. Int. Ed. Engl. 33 (1994) 1775e1778.
[32] T. Koizumi, T. Tomon, K. Tanaka, Organometallics (2003) 970e975.
[33] C. Moorlag, M.O. Wolf, C. Bohne, B.O. Patrick, J. Am. Chem. Soc. 127 (2005)
6382e6393.
[34] N. Vargas-Alfredo, N. Espinosa-Jalapa, S. Lopez-Morales, A.D. Ryabov, R. Le
Lagadec, L. Alexandrova, Macromolecules 45 (2012) 8135e8146.
[35] N. Vargas-Alfredo, C. Aguilar-Lugo, O. Gonzalez-Diaz, R. Le Lagadec,
L. Alexandrova, Macromol. Symp. 325e326 (2013) 10e20.
Acknowledgments
Financial support from CONACyT (Projects 153151 and 129801,
[36] M.O. Gonzalez-Diaz, S. Lopez-Morales, R. Le Lagadec, L. Alexandrova, J. Polym.

310
V. Martínez Cornejo et al. / Journal of Organometallic Chemistry 799-800 (2015) 299e310
Sci. Part A Polym. Chem. 49 (2011) 4562e4577.
[37] J. Olvera-Mancilla, S. Lopez-Morales, J. Palacios-Alquisira, D. Morales-Morales,
R. Le Lagadec, L. Alexandrova, Polymer 55 (2014) 1656e1665.
B.M. Stoltz, J.E. Bercaw, K.I. Goldberg, Organometallics 29 (2010) 2176e2179.
[47] J.G. Małecki, M. Jaworska, R. Kruszynski, J. Kłak, Polyhedron 24 (2005)
3012e3021.
ꢀ
[38] L. Quebatte, R. Scopelliti, K. Severin, Eur. J. Inorg. Chem. 2 (2005) 3353e3358.
[39] B. De Clercq, F. Verpoort, Polym. Bull. 50 (2003) 153e160.
[40] O. Saavedra-Díaz, R. Ceron-Camacho, S. Hernandez, A.D. Ryabov, R. Le Laga-
dec, Eur. J. Inorg. Chem. (2008) 4866e4869.
[41] R.A. Zelonka, M.C. Baird, Can. J. Chem. 50 (1972) 3063e3072.
[42] R.A. Zelonka, M.C. Baird, J. Organomet. Chem. 44 (1972) 383e389.
[43] AXS, SAINT, Software Reference Manual, Bruker, Madison, WI, 1998.
[44] G.M. Sheldrick, SHELXS-2012 and SHELXL-2014/7, University of Gottingen,
Germany.
[48] M. Kamigaito, T. Ando, M. Sawamoto, Chem. Rev. 101 (2001) 3689e3745.
[49] G. Moad, D.H. Solomon, The Chemistry of Radical Polymerization, Elsevier,
2005.
[50] M. Wakioka, K.-Y. Baek, T. Ando, M. Kamigaito, M. Sawamoto, Macromolecules
35 (2002) 330e333.
[51] M.A. Bennett, H. Neumann, M. Thomas, X. Wang, Organometallics 10 (1991)
3237e3245.
[52] D.W. Lee, C.S. Yi, Organometallics 29 (2010) 3413e3417.
[53] M.A. Bennett, T.-N. Huang, T.W. Matheson, A.K. Smith, Inorg. Synth. 21 (1982)
74.
ꢀ
ꢀ
[45] W. Levason, G. Reid, Coord. Chem. Rev. 250 (2006) 2565e2594.
[46] G.R. Fulmer, A.J.M. Miller, N.H. Sherden, H.E. Gottlieb, A. Nudelman,
[54] C.A. Tolman, Chem. Rev. 77 (1977) 313e348.