Synthesis and characterization of a family of Co(II) triphenylamido-amine complexes and catalytic activity in controlled radical polymerization of olefins

Article abstract of DOI:10.1016/j.poly.2012.11.020

The present work reports a new family of tripodal Co^II complexes bearing trianionic triphenylamido-amine ligands with a variety of pendant arms (aryl, acyl, alkyl). These complexes have been synthesized by the reaction of anhydrous CoCl₂ with the deprotonated ligands and exhibit stoichiometric and structural variation. The solid-state structures of these compounds reveal that in all cases the four nitrogen-atom residues of the ligands are coordinated to the metal center in a distorted trigonal-pyramidal geometry. In two cases, the presence of acetonitrile adds a fifth moiety to the coordination sphere. Among the aryl-armed Co^II reagents, [K(L ³)Co^II-NCMe]_n (1), [K(THF)₆][(L ⁵)Co^II]·1.5THF (2) and [K(NCMe)₃(L ¹³)Co^II-NCMe] (6), the two five-coordinate structures (1, 6) exhibit distorted trigonal bipyramidal geometries, with that of 6 being the least distorted. The four-coordinate species 2 is the only compound with a distinct anionic [(L⁵)Co^II]^- component, since the K⁺ ion is solely coordinated by solvent molecules. The acyl-armed Co^II compounds [K(THF)₂(L⁸)Co ^II]_n (3) and {[K₂(DMA)₃(L ¹⁰)₂Co^II₂]·0.5Et ₂O}_n (5) are strictly four-coordinate species, with the carbonyl moieties oriented exo with respect to the cavity of the vacant coordination site. Finally, compound [K₂(L⁹) ₂Co^II₂]_n (4) is the only example bearing an alkyl-armed ligand, and exhibits a geometry featuring a repeating -[Co(1)-K(1)-Co(2)-K(2)]_n- sequence. All compounds have been characterized by spectroscopic and electrochemical techniques. Compounds 1, 2, 4, and 6 show reversible or semi-reversible Co^II/Co^III redox couples, whereas the electron-deficient complexes 3 and 5 exhibit irreversible anodic waves. The catalytic reactivity of these complexes towards controlled radical polymerization (CRP) of styrene (St) and methyl methacrylate (MMA) has been studied and preliminary results are presented. Compounds 4 and 5 seem to be the most reactive with both monomers, giving high yields of polymers (60-86%), 3 efficiently induces styrene polymerization (90%), whereas 2 and 6 provide lower yields (12-16% for PS and 30-45% for PMMA). The steric factor seems to play an essential role, since Co^II compounds that feature a less hindered fifth coordination site show the highest reactivity and better control over polymerization. The polymers obtained are predominately syndiotactic, consistent with radical polymerization, with two exceptions (1, 2) that exhibit unusually high ratios of isotactic triads. Interestingly, in many cases, the polymers obtained feature bimodal distributions, while the molecular weight distributions are not very broad (1.40-2.00), and this strongly indicates that two parallel mechanisms may be in operation.

Full text of DOI:10.1016/j.poly.2012.11.020

Polyhedron 52 (2013) 78–90
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Synthesis and characterization of a family of Co(II) triphenylamido-amine
complexes and catalytic activity in controlled radical polymerization of olefins
Vivek Bagchi a, Grigorios Raptopoulos b, Purak Das a, Sotirios Christodoulou b, Qiuwen Wang a, Lin Ai a,
Amitava Choudhury a, Marinos Pitsikalis c, Patrina Paraskevopoulou b, , Pericles Stavropoulos a,
⇑
⇑
a Department of Chemistry, Missouri University of Science and Technology, Rolla, MO 65409, USA
b Inorganic Chemistry Laboratory, Department of Chemistry, University of Athens, Panepistimiopolis Zografou 15771, Athens, Greece
c Industrial Chemistry Laboratory, Department of Chemistry, University of Athens, Panepistimiopolis Zografou 15771, Athens, Greece
a r t i c l e i n f o
a b s t r a c t
The present work reports a new family of tripodal CoII complexes bearing trianionic triphenylamido-
amine ligands with a variety of pendant arms (aryl, acyl, alkyl). These complexes have been synthesized
by the reaction of anhydrous CoCl2 with the deprotonated ligands and exhibit stoichiometric and
structural variation. The solid-state structures of these compounds reveal that in all cases the four nitro-
gen-atom residues of the ligands are coordinated to the metal center in a distorted trigonal-pyramidal
geometry. In two cases, the presence of acetonitrile adds a fifth moiety to the coordination sphere. Among
Article history:
Available online 29 November 2012
Dedicated to the memory of Alfred Werner,
the founder of modern coordination
chemistry, on the 100th Anniversary of his
Nobel Prize in Chemistry in 1913.
the aryl-armed CoII reagents, [K(L3)CoII–NCMe]n (1), [K(THF)6][(L5)CoII]ꢀ1.5THF (2) and [K(NCMe)3(L13
)
CoII–NCMe] (6), the two five-coordinate structures (1, 6) exhibit distorted trigonal bipyramidal geome-
tries, with that of 6 being the least distorted. The four-coordinate species 2 is the only compound with
a distinct anionic [(L5)CoII]ꢁ component, since the K+ ion is solely coordinated by solvent molecules.
The acyl-armed CoII compounds [K(THF)2(L8)CoII]n (3) and {[K2(DMA)3(L10)2CoII2]ꢀ0.5Et2O}n (5) are strictly
four-coordinate species, with the carbonyl moieties oriented exo with respect to the cavity of the vacant
Keywords:
Cobalt complexes
Trisamidoamine ligands
Electrochemistry
Controlled radical polymerization
Catalysis
coordination site. Finally, compound [K2(L9)2CoII
]n (4) is the only example bearing an alkyl-armed ligand,
2
Atom-transfer reactions
and exhibits a geometry featuring a repeating –[Co(1)–K(1)–Co(2)–K(2)]n– sequence. All compounds
have been characterized by spectroscopic and electrochemical techniques. Compounds 1, 2, 4, and 6 show
reversible or semi-reversible CoII/CoIII redox couples, whereas the electron-deficient complexes 3 and 5
exhibit irreversible anodic waves. The catalytic reactivity of these complexes towards controlled radical
polymerization (CRP) of styrene (St) and methyl methacrylate (MMA) has been studied and preliminary
results are presented. Compounds 4 and 5 seem to be the most reactive with both monomers, giving high
yields of polymers (60–86%), 3 efficiently induces styrene polymerization (90%), whereas 2 and 6 provide
lower yields (12–16% for PS and 30–45% for PMMA). The steric factor seems to play an essential role, since
CoII compounds that feature a less hindered fifth coordination site show the highest reactivity and better
control over polymerization. The polymers obtained are predominately syndiotactic, consistent with
radical polymerization, with two exceptions (1, 2) that exhibit unusually high ratios of isotactic triads.
Interestingly, in many cases, the polymers obtained feature bimodal distributions, while the molecular
weight distributions are not very broad (1.40–2.00), and this strongly indicates that two parallel
mechanisms may be in operation.
Ó 2012 Elsevier Ltd. All rights reserved.
1. Introduction
[3]. More recent work has shown that cobalt reagents can be
intriguingly more complex in their mode of operation, and can in-
Radical polymerization reactions mediated by cobalt reagents
[1] have played a historic role in the development of controlled
radical polymerization (CRP) processes, owing to the early, seminal
contributions of Wayland et al. [2] and Harwood and co-workers
duce CRP reactions not only via atom-transfer radical polymeriza-
tion (ATRP) mechanisms, but also via organometallic-mediated
radical polymerization (OMRP) pathways that rely on the well-
known propensity of the Co–C bond towards cleavage and genera-
tion of carbon-centered radicals.
For the mechanistically related atom-transfer radical addition
(ATRA) and polymerization (ATRP) reactions, the key step is a
reversible halogen-atom transfer between an organohalide (RX)
and the cobalt site, shuttling between two oxidation states
⇑
Corresponding authors. Tel.: +30 210 727 4381; fax: +30 210 727 4782 (P.
Paraskevopoulou), tel.: +1 573 341 7220; fax: +1 573 341 6033 (P. Stavropoulos).
E-mail addresses: paraskevopoulou@chem.uoa.gr (P. Paraskevopoulou),
pericles@mst.edu (P. Stavropoulos).
0277-5387/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2012.11.020

V. Bagchi et al. / Polyhedron 52 (2013) 78–90
79
(CoII/CoIII) (Chart 1) [4,5]. The resulting RÅ radical could then add to
an olefin to generate a new organic radical, which would either be
halogenated to provide the monomeric product (ATRA) [6] or con-
tinue adding olefin to afford a living polymer (ATRP). Its success re-
lies on the control exercised by the dynamic equilibrium
(KATRP = kact/kdeact) established between the halide-capped dor-
consisting of CoII immobilized on an ion exchange resin along with
a small amount of soluble catalyst (CuCl2/Me6TREN) [14]. Among
CoI systems, [CoX(PPh3)3] (X = Cl, Br, I) are the first compounds re-
ported to mediate ATRP reactions [15], with the iodide being the
most reactive. On the other hand, the OMRP mechanism is report-
edly obeyed by [Co(acac)2] in the polymerization of vinyl acetate,
and co-polymerization of vinyl acetate and vinyl chloroacetate
[16–19], as well as by CoII porphyrin complexes in the polymeriza-
tion of acrylate monomers [2,20–22]. Finally, low-spin CoII
complexes, such as cobaloximes [23], CoII-5,10,15,20-tetraphenyl-
21H,23H-porphine [24], and CoII-glyoximato species [25–28]
operate via the CCTP mechanism in the (co)polymerization of
Å
mant species (Pn–X) and the propagating radical (Pn ), which is very
sensitive to the catalyst, olefin monomer, and RX (initiator) used,
and depend on the solvent and the reaction conditions [7].
On the other hand, the OMRP process (or CMRP, for cobalt-med-
iated radical polymerization) relies on an equilibrium (Chart 2(a),
demonstrated by vinyl acetate) heavily favoring a cobalt-capped
dormant form of the polymer radical (CoIII–Pn). An alternative, al-
beit related, process, known as catalytic chain transfer polymeriza-
tion (CCTP), applies to substrates that are prone to b-H elimination
(meth)acrylates,
a-methylstyrene and styrenes, yielding low
molecular weight macromolecules in organic and aqueous media.
Recently, we, and others, have been exploring the synthesis and
reactivity of a series of tripodal transition metal compounds
(M = Fe, Mn, Co) with rigid triphenylamido-amine cores and a vari-
ety of pendant arms (R = aryl, acyl, alkyl) (Chart 3) [29–38]. While
we have been largely investigating catalytic C–O and C–N bond
construction via oxo- and nitrene-transfer chemistry, we have
more recently recognized that many members of this family of re-
agents are active in CRP catalysis, enabled by a reversible or semi-
reversible Mn+/M(n+1)+ couple at easily accessible, and frequently
highly reducing potentials. In this manuscript we are reporting
the synthesis and characterization of a series of tripodal CoII triph-
enylamido-amine complexes bearing a selection of aryl, acyl and
alkyl arms. We are also disclosing CRP related data, mediated by
these reagents.
(methacrylates,
a-methylstyrene, methacrylonitrile), inasmuch as
the CoIII–Pn species may induce b-H atom abstraction to generate
a CoIII–H and a polymer chain with an unsaturated terminal group
(Chart 2(b), demonstrated for methyl methacrylate (MMA)). The
CoIII–H is a key ingredient of the catalytic cycle, because it can gen-
erate the primary radical species via insertion of the hydride into
the monomer (for instance MMA).
The ATRP mechanism is apparently followed by cobaltocene,
acting as a precatalyst in the polymerization of styrene (St) and
MMA [8]. Other common CoII species, such as cobalt carboxylates
are also known to induce ATRP reactions of MMA and St (oligomer-
ization) [9–11]. In particular, cobalt acetate polymerizes MMA in
good yields (82%) and narrow polydispersity (Mw/Mn = 1.26). Addi-
tion of small amounts of CoCl2, Cu(OAc)2 or tris(2-(dimethyl ami-
no) ethyl)amine (Me6TREN), improves the rate of the reaction,
but the control over the polymerization is reduced (Mw/
Mn = 1.49–1.74) [9]. The CoCl2/Me6TREN system has been reported
as catalyst for the ATRP of MMA, exerting modest control over the
polymerization and molecular weight distribution of PMMA (Mw/
Mn = 1.63–1.80) [12], which can be improved by addition of a small
amount of hybrid deactivators (FeBr3/Me6TREN or CuBr2/Me6-
TREN; Mw/Mn = 1.15–1.46). Heterogeneous versions have also been
developed, such as a CoII/CuII bimetallic catalyst immobilized on a
cross-linked poly(acrylic acid) resin [13] or a ‘‘hybrid’’ catalyst,
2. Experimental
2.1. Materials and physical measurements
All operations were performed under anaerobic conditions un-
der a pure dinitrogen or argon atmosphere using Schlenk tech-
niques on an inert gas/vacuum manifold or in a dry-box (O2,
H2O < 1 ppm). Anhydrous diethyl ether, acetonitrile, tetrahydrofu-
ran, hexane, pentane, toluene, chloroform, dimethylformamide
(DMF), and dimethylacetamide (DMA) were purchased from Sig-
ma–Aldrich. Methanol was distilled over the corresponding mag-
nesium alkoxide, and acetone was distilled over drierite. Solvents
were degassed by three freeze–pump–thaw cycles. Unless other-
wise noted, all other reagents were purchased at the highest purity
available. Potassium hydride was provided as dispersion in mineral
oil and was thoroughly washed prior to use with copious amounts
of tetrahydrofuran followed by hexane. Compounds L3H3, L5H3,
L8H3 and L9H3 have been prepared according to literature methods
[31].
1H NMR and 13C NMR spectra were recorded on a Varian XL-
400, a Varian INOVA/UNITY 400 MHz and a Varian 300 Unity Plus
NMR spectrometers. IR spectra were obtained on a Perkin-Elmer
883 IR spectrometer and FT-IR spectra on a Shimadzu IR-Affinity-
1 spectrometer. UV–Vis spectra were obtained on a Hewlett–Pack-
ard 8452A diode array, a Varian Cary 50, and a Varian Cary 300
spectrophotometers. Microanalyses were done by Galbraith Labo-
ratories, Knoxville, TN, Quantitative Technologies Inc., Whitehouse,
NJ, and on an in-house Perkin-Elmer 2400 CHN analyzer. HRMS
data were collected on a Thermo Fisher Scientific LTQ-Orbitrap
XL hybrid mass spectrometer, using the Orbitrap analyzer for
acquisition of high-resolution accurate mass data. Samples were
infused using the integrated syringe pump at 3 lL/min and ioniza-
tion was via the electrospray source with source settings at their
defaults. In general, settings for the ion optics were determined
automatically during the regular tuning and calibration of the
instrument. For high-resolution data, the Orbitrap analyzer is set
Chart 1.

80
V. Bagchi et al. / Polyhedron 52 (2013) 78–90
Chart 2.
1000 mV/s. All potentials are reported versus the ferrocenium/fer-
rocene (Fc+/Fc) couple.
2.2. Synthesis
2.2.1. Synthesis of [(tert-butyl-O-CO)-NH-(2-C6H4)]3N (L10H3)
In a 100 mL Schlenk flask was added (2-NH2-C6H4)3N (1.16 g,
4.0 mmol) and NaH (0.43 g, 18.0 mmol) in THF (30 mL), and the
mixture was refluxed for 2 h. The reaction mixture was allowed
to cool at room temperature, di-tert-butyldicarbonate (3.45 mL,
15.0 mmol) was added, and then the mixture was refluxed for
12 h. Water (5.0 mL) was then added to the reaction mixture, fol-
lowed by further refluxing for 6 h. The solvent was removed, and
the crude residue was dissolved in dichloromethane. This solution
was dried over Na2SO4. The product was purified by column chro-
matography (silica gel), and the fragment eluting with petroleum
ether/ethyl acetate (8:1) was collected to give the ligand as a white
crystalline solid (1.94 g, 82%). 1H NMR (CDCl3, 7.27 ppm): d 7.58 (d,
3H, aryl), 7.08 (t, 3H, aryl), 7.03 (s, 3H, NH), 6.95 (t, 3H, aryl), 6.74
(d, 3H, aryl), 1.32 (s, 27H). 13C NMR (CDCl3, 77 ppm): d 153.28,
138.18, 131.93, 124.88, 124.50, 124.18, 79.77, 28.09. IR (KBr,
cmꢁ1): 3427, 3360, 3239, 3119, 2980, 2931, 1734, 1692, 1595,
1528, 1476, 1443, 1391, 1366, 1300, 1277, 1239, 1152, 1052,
1025, 942, 923, 857, 757, 619. HRMS calcd. for C33H42N4O6 (M+):
590.30955; found 590.31183. Elem. Anal. Calc. for C33H42N4O6: C,
67.10; H, 7.17; N, 9.48. Found: C, 67.14; H, 7.23; N, 9.53%.
Chart 3.
to a resolution of 100000. SEC experiments were carried out at
40 °C using a modular instrument consisting of a Waters model
510 pump, U6K sample injector, 401 refractometer, 486 UV spec-
trophotometer, and a set of 5 l-Styragel columns with a continuous
porosity range from 500 to 106 Å. The columns were housed in an
oven thermostated at 40 °C. CHCl3 was the carrier solvent at a flow
rate of 1 mL/min. The system was calibrated with nine PS stan-
dards having molecular weights in the range of 970–600,000.
Cyclic voltammetry was carried out with an Eco Chemie Auto-
lab PGSTAT100 electrochemical workstation fitted in
a Dry
Box and controlled with a General Purpose Electrochemical Soft-
ware (GPES) or with a Bipotentiostat AFCBP1 from Pine Instrument
Company fitted in a Dry Box and controlled with the PineChem
2.7.9 software. Experiments were performed using a gold disk
working electrode (2 mm diameter) and a Ag/Ag+ (0.01 M AgNO3
and 0.5 M [(n-Bu)4N]PF6 in MeCN or DMF) non-aqueous reference
electrode (Bioanalytical Systems, Inc.) with a prolonged bridge
(0.5 M [(n-Bu)4N]PF6 in MeCN or DMF). A thin Pt foil or gauge
(8 cm2, Sigma–Aldrich) was employed as counter electrode. The
2.2.2. Synthesis of [(2,6-di-methyl-C6H3)-NH-(2-C6H4)]3N (L13H3)
In a 100 mL Schlenk flask, [Pd2(dba)3] (tris(dibenzylideneace-
tone)dipalladium, 0.0912 g, 0.10 mmol) and BINAP (2,20-
bis(diphenylphosphino)-1,10-binaphthyl, 0.1866 g, 0.30 mmol)
were stirred in anhydrous toluene (10 mL), and heated until BINAP
dissolved. To the resulting cherry-red solution was added (2-NH2-
C6H4)3N (0.58 g, 2.0 mmol), 2,6-dimethylbromobenzene (2.22 g,
12.0 mmol), and tert-Bu-ONa (1.96 g, 20.0 mmol). The slurry was
refluxed for 48 h to give a brown solution, with precipitation of a
solid. The mixture was allowed to cool to room temperature, the
solution was filtered, and the filtrate was evaporated under vac-
uum to give a brown residue. The residue was purified by column
chromatography (silica gel). The fragment eluting with petroleum
working electrode was polished using successively 6, 3, 1 lm dia-
mond paste on a DP-Nap polishing cloth (Struers, Westlake, OH),
washed with water, acetone and air-dried. The Pt foil and gauge
electrodes were cleaned in a H2O2/H2SO4(conc) solution (1:4 v:v)
and oven-dried. The concentrations of the samples were between
1 and 3 mM and that of [(n-Bu)4N]PF6 (supporting electrolyte)
was 0.5 M. The potential sweep rate varied between 10 and

V. Bagchi et al. / Polyhedron 52 (2013) 78–90
81
ether/ethyl acetate (50:1 v/v) was collected, and the resulting pale
712, 665, 622, 562, 552, 498, 463. UV–Vis (THF, nm), (
277 (sh), 464 (sh), 578 (30); (DMF, nm), (e
576 (243). Elem. Anal. Calc. for C32H28N4F9O5CoK: C, 46.99; H, 3.45;
N, 6.85. Found: C, 47.09; H, 3.49; N, 6.83%.
e
, Mꢁ1 cmꢁ1):
yellow solid was recrystallized from THF/petroleum ether to afford
the ligand as a crystalline solid suitable for X-ray diffraction anal-
ysis (0.77 g, 64%). 1H NMR (CDCl3, 7.27 ppm): d 7.19 (d, 3H, aryl),
6.93 (m, 12H, aryl), 6.73 (t, 3H, aryl), 6.16 (d, 3H, aryl), 5.6 (s, 3H,
NH), 2.22 (s, 9H), 1.8 (s, 9H), 1.52 (s, 9H). 13C NMR (CDCl3,
, Mꢁ1 cmꢁ1): 540 (200),
2.2.6. Synthesis of [K2(L9)2CoII
] (4)
2 n
77 ppm):
d
141.08, 137.88, 136.05, 135.81, 132.22, 128.61,
The ligand L9H3 (0.208 g, 0.5 mmol) was dissolved in degassed
THF (15.0 mL), and KH (0.060 g, 1.5 mmol) was added to this solu-
tion. The mixture was allowed to stir overnight, followed by addi-
tion of anhydrous CoCl2 (0.065 g, 0.5 mmol). This mixture was
stirred for an additional 12 h to afford a dark purple solution. The
solution was refrigerated (ꢁ30 °C) overnight and the insoluble
salts were filtered off on an anaerobic frit. The filtrate was reduced
to 3.0 mL by evaporation, and was allowed to stand at room
temperature to afford dark purple crystalline material, suitable
for X-ray diffraction analysis (0.218 g, 85%). FT-IR (KBr, cmꢁ1):
3399, 3046, 2963, 2866, 1654, 1599, 1587, 1560, 1541, 1508,
1478, 1383, 1326, 1303, 1262, 1235, 1173, 1150, 1117, 1045,
128.46, 126.21, 125.91, 118.55, 112.75, 17.99, 17.41. IR (KBr,
cmꢁ1): 3378, 3342, 3064, 3032, 2948, 2916, 2851, 1925, 1586,
1499, 1473, 1454, 1307, 1263, 1225, 1153, 1099, 1036, 916, 770,
745, 692, 623. HRMS calcd. for C42H42N4 (M+): 602.33983; found
602.34180. Elem. Anal. Calc. for C42H42N4: C, 83.68; H, 7.02; N,
9.29. Found: C, 83.54; H, 7.11; N, 9.27%.
2.2.3. Synthesis of [K(L3)CoII–NCMe]n (1)
The ligand L3H3 (0.463 g, 0.5 mmol) was dissolved in degassed
THF (15.0 mL), and KH (0.060 g, 1.5 mmol) was added to this solu-
tion. The mixture was allowed to stir overnight, followed by addi-
tion of anhydrous CoCl2 (0.065 g, 0.5 mmol). This mixture was
stirred for an additional 24 h to afford a dark brown solution. The
solution was refrigerated (ꢁ30 °C) for about 6 h. The insoluble salts
were filtered off on an anaerobic frit, the filtrate was evaporated to
dryness, and the residue was dissolved in acetonitrile (3.0 mL).
Diethyl ether (10.0 mL) was carefully layered over acetonitrile,
and pentane (10.0 mL) was further carefully layered over the ether.
The system is allowed to slowly mix at room temperature to afford
brown crystalline material of the title compound, suitable for X-ray
diffraction analysis (0.383 g, 72%). IR (KBr, cmꢁ1): 3435, 3043,
2974, 2873, 1610, 1587, 1482, 1466, 1388, 1276, 1171, 1127,
1047, 996, 965, 859, 747, 702, 682, 668, 614, 545, 486, 425. UV–
987, 894, 821, 741, 627, 536. UV–Vis (DMF, nm), (e
, Mꢁ1 cmꢁ1):
598 (700), 619 (sh). Elem. Anal. Calc. for C27H33N4CoK: C, 63.37;
H, 6.50; N, 10.95. Found: C, 63.44; H, 6.41; N, 10.86%.
2.2.7. Synthesis of {[K2(DMA)3(L10)2CoII2]ꢀ0.5Et2O}n (5)
The ligand L10H3 (0.295 g, 0.5 mmol) was dissolved in degassed
DMA (5.0 mL), and KH (0.060 g, 1.5 mmol) was added to this solu-
tion. The mixture was allowed to stir overnight, followed by addi-
tion of anhydrous CoCl2 (0.065 g, 0.5 mmol). This mixture was
stirred for an additional 24 h to afford a blue-violet solution. The
solution was refrigerated (ꢁ30 °C) overnight, and the insoluble
salts were filtered off on an anaerobic frit. Diethyl ether (20 mL)
was carefully layered over the DMA filtrate, and pentane (10 mL)
was layered over the diethyl ether. The setup was allowed to stand
at room temperature to afford pink-violet crystalline material,
suitable for X-ray diffraction analysis (0.250 g, 59%). FT-IR (KBr,
cmꢁ1): 3448, 3061, 2977, 2929, 1635, 1587, 1485, 1449, 1388,
1363, 1239, 1177, 1113, 1050, 1027, 939, 842, 780, 751, 621,
Vis (MeCN, nm), (e
, Mꢁ1 cmꢁ1): 338 (33100), 473 (sh), 575 (sh),
745 (110). Elem. Anal. Calc. for C44H24N5F18CoK: C, 49.71; H,
2.28; N, 6.59. Found: C, 49.61; H, 2.31; N, 6.64%.
2.2.4. Synthesis of [K(THF)6][(L5)CoII]ꢀ1.5THF (2)
The ligand L5H3 (0.363 g, 0.5 mmol) was dissolved in degassed
THF (15.0 mL), and KH (0.060 g, 1.5 mmol) was added to this
solution. The mixture was allowed to stir overnight, followed by
addition of anhydrous CoCl2 (0.065 g, 0.5 mmol). This mixture
was stirred for an additional 24 h to afford a green-brown solution.
The solution was refrigerated (ꢁ30 °C) overnight and the insoluble
salts were filtered off on an anaerobic frit. The filtrate was reduced
to 3.0 mL by evaporation, and pentane (10.0 mL) was carefully lay-
ered over the THF solution. The system is allowed to slowly mix at
room temperature to afford green crystalline material, suitable for
X-ray diffraction analysis (0.386 g, 57%). FT-IR (KBr, cmꢁ1): 3448,
3097, 3057, 2974, 2872, 1560, 1540, 1534, 1478, 1438, 1330,
1283, 1220, 1157, 1103, 1048, 981, 963, 891, 857, 826, 788, 753,
592, 493, 475, 453, 425. UV–Vis (DMA, nm), (
e
, Mꢁ1 cmꢁ1): no
peaks; (MeCN, nm), (
e
, Mꢁ1 cmꢁ1): 270 (sh), 535 (54). Elem. Anal.
Calc. for C41H57.5N5.5O8CoK: C, 57.23; H, 6.74; N, 8.95. Found: C,
57.59; H, 6.74; N, 9.11%.
2.2.8. Synthesis of [K(NCMe)3(L13)CoII–NCMe] (6)
The ligand L13H3 (0.301 g, 0.5 mmol) was dissolved in degassed
THF (15.0 mL), and KH (0.060 g, 1.5 mmol) was added to this solu-
tion. The mixture was allowed to stir overnight, followed by addi-
tion of anhydrous CoCl2 (0.065 g, 0.5 mmol). This mixture was
stirred for an additional 24 h to afford a deep green-brown solu-
tion. The solution was refrigerated (ꢁ30 °C) overnight and the
insoluble salts were filtered off on an anaerobic frit. The filtrate
was evaporated to dryness, and the residue was dissolved in aceto-
nitrile (10.0 mL). This solution was reduced to approximately 5 mL
by evaporation, and layered first with diethyl ether (10 mL) and
then with pentane (5 mL) to afford green-brown crystalline mate-
rial, suitable for X-ray diffraction analysis (0.297 g, 69%). FT-IR
(KBr, cmꢁ1): 3446, 3386, 3224, 3045, 2916, 2257, 2125, 1636,
1592, 1582, 1560, 1507, 1479, 1443, 1336, 1302, 1207, 1144,
1091, 965, 916, 854, 829, 766, 736, 625, 524, 354. UV–Vis (MeCN,
747, 669, 620, 594, 583, 547, 489. UV–Vis (THF, nm), (
420 (sh), 457 (sh), 606 (160); (DMF, nm), (
(1900), 556 (2000). Elem. Anal. Calc. for C66H81N4Cl6O7.5CoK: C,
58.27; H, 6.00; N, 4.12. Found: C, 58.37; H, 6.09; N, 4.23%.
e
, Mꢁ1 cmꢁ1):
e
, Mꢁ1 cmꢁ1): 528
2.2.5. Synthesis of [K(THF)2(L8)CoII]n (3)
The ligand L8H3 (0.289 g, 0.5 mmol) was dissolved in degassed
THF (15.0 mL), and KH (0.060 g, 1.5 mmol) was added to this
solution. The mixture was allowed to stir overnight, followed by
addition of anhydrous CoCl2 (0.065 g, 0.5 mmol). This mixture
was stirred for an additional 12 h to afford a dark purple solution.
The solution was refrigerated (ꢁ30 °C) overnight and the insoluble
salts were filtered off on an anaerobic frit. The filtrate was reduced
to 3.0 mL by evaporation, and was allowed to stand at room tem-
perature to afford dark purple crystalline material, suitable for X-
ray diffraction analysis (0.368 g, 90%). FT-IR (KBr, cmꢁ1): 3434,
3067, 2974, 2873, 1679, 1646, 1592, 1481, 1453, 1420, 1254,
1207, 1161, 1142, 1055, 953, 946, 910, 826, 773, 760, 748, 727,
nm), (e
, Mꢁ1 cmꢁ1): 254 (36500), 370 (sh), 459 (1700), 542 (sh).
Elem. Anal. Calc. for C50H51N8CoK: C, 69.65; H, 5.97; N, 13.00.
Found: C, 69.63; H, 5.90; N, 12.89%.
2.3. X-ray crystallography
Intensity data sets for compounds 1–6 were collected on a
Bruker Smart Apex diffractometer using SMART software [39]. Suit-
able crystals were selected and mounted on a glass fiber using

82
V. Bagchi et al. / Polyhedron 52 (2013) 78–90
Table 1
Crystallographic Data for CoII compounds 1–6.
Crystallographic parameters
1
2
3
4
5
6
Formula
Mr
Color
Crystal system
Space group
a (Å)
C
44H24CoF18KN5
C66H81Cl6CoKN4O7.5
1361.08
green
monoclinic
C2
23.804(3)
13.7687(17)
21.403(3)
90
105.120(2)
90
6771.8(2)
4
1.335
173.2(4)
0.71073
0.606
C32H28CoF9KN4O5
817.61
purple
monoclinic
P21
9.9316(10)
17.0639(17)
10.3409(11)
90
101.872(2)
90
1715.0(3)
2
1.583
130(2)
C54H66Co2K2N8
1023.21
green
C82H115Co2K2N11O16
1706.91
light blue
C50H51CoKN8
862.02
light brown
1062.71
light brown
monoclinic
P21/n
14.2383(16)
13.5103(15)
21.573(2)
90
93.938(2)
90
4140.0(8)
4
1.705
140(2)
0.71073
0.638
0.0524
triclinic
triclinic
triclinic
ꢀ
ꢀ
ꢀ
P1
P1
P1
9.520(2)
13.186(3)
20.212(4)
88.959(3)
80.804(3)
85.954(4)
2498.5(9)
2
1.360
130(2)
0.71073
0.876
0.0861
10.4909(12)
19.816(2)
21.881(3)
83.0470(10)
88.0360(10)
81.2780(10)
4462.6(9)
2
1.270
136(2)
0.71073
0.532
0.0409
12.3029(13)
12.4441(14)
14.9916(16)
86.1900(10)
81.9630(10)
86.5190(10)
2264.6(4)
2
1.264
136(2)
0.71073
0.514
0.0370
b (Å)
c (Å)
a
(°)
b (°)
c
(°)
V (Å3)
Z
Dcalc (mg mꢁ3
T (K)
)
k (Å)
0.71073
0.717
0.0336
l
(mmꢁ1
)
r
a
R1 (I > 2
wR2 (I > 2
(I))
(I))
0.0582
0.1466
b
r
0.1157
0.0759
0.2094
0.1083
0.1007
a
R1
wR2 = [
=
R
||Fo| ꢁ |Fc||/
R|Fo|.
b
2
R
w(Fo ꢁ Fc2)2/
R .
w(Fo2)2]1/2
epoxy-based glue. The data sets were collected at low tempera-
tures employing a scan of 0.3° in with an exposure time of
2.4.2. Reinitiation experiment 1
x
Compound [K2(L9)2CoII2]n (4) (10.2 mg, 0.01 mmol) and the sol-
vent (toluene, 2.0 mL) were added to a Schlenk flask, followed by St
(1.0 g, 9.6 mmol) and the initiator (ethyl-2-bromo-isobutyrate,
20 s/frame. The cell refinement and data reduction were carried
out with SAINT [40], the program SADABS was used for the absorption
correction. The structure was solved by direct methods using SHEL-
XS-97 [41] and difference Fourier syntheses. Full-matrix least-
squares refinement against |F2| was carried out using the SHELXTL-
PLUS suit of programs. All non-hydrogen atoms were refined aniso-
tropically. Hydrogen atoms were placed geometrically and held in
the riding mode during the final refinement. In compound 5, there
were disorders in the solvent molecules, for example two dimethyl
acetamide molecules existed in two different orientations and two
carbon atoms of the diethyl ether had large thermal parameters.
The ratios of the two different orientations were obtained by freely
refining the occupancies of the atoms, which revealed the ratio of
50:50 for one DMA and 70:30 for the other. The carbon atoms with
large thermal parameters for the ether were also divided into two
positions and their occupancies were refined freely. Soft restraints
were used to keep bond lengths within reasonable values. Perti-
nent crystallographic data are collected in Tables 1 and S1. Selected
bond distances and angles for compounds 1–6 are given in Tables 2
and 3, respectively.
7.5
l
L, 5.1 ꢂ 10ꢁ5 mol). The flask was immediately immersed in
the oil bath at 110 °C and kept for 20 h. After heating was stopped,
the reaction vessel was cooled to room temperature and an aliquot
(0.5 mL) was taken. To this aliquot a large excess of MeOH was
added to precipitate the polymer formed, which was purified and
characterized as described above.
To the rest of the reaction mixture, St (1.0 g, 9.6 mmol) was
added and was left to react for another 20 h at 110 °C. After heating
was stopped, the reaction vessel was cooled to room temperature
and a large excess of MeOH was added to precipitate the polymer
formed, which was purified and characterized as described above.
2.4.3. Reinitiation experiment 2
Compound [K(THF)6][(L5)CoII]ꢀ1.5THF (2) (13.6 mg, 0.01 mmol)
and the solvent (toluene, 2.0 mL) were added to a Schlenk flask,
followed by MMA (1.0 g, 10.0 mmol) and the initiator (ethyl-2-bro-
mo-isobutyrate, 7.5
l
L, 5.1 ꢂ 10ꢁ5 mol). The flask was immediately
immersed in the oil bath at 110 °C and kept for 16 h. After heating
was stopped, the reaction vessel was cooled to room temperature
and a large excess of MeOH was added to precipitate the polymer
formed, which was purified and characterized as described above.
The PMMA obtained was redissolved in toluene (2.0 mL), where
the metal complex (13.6 mg, 0.01 mmol) and St (1.0 g, 9.6 mmol)
were added and the mixture was left to react for another 20 h at
110 °C. After heating was stopped, the reaction vessel was cooled
to room temperature and a large excess of MeOH was added to pre-
cipitate the polymer formed, which was purified and characterized
as described above.
2.4. Polymerization reactions
2.4.1. Typical experimental procedure
All operations were carried out under an inert atmosphere. The
CoII compound (0.01 mmol; e.g. [K(L3)CoII–NCMe]n (1): 10.6 mg)
and the solvent (toluene, 1.0 mL) were added to a Schlenk flask, fol-
lowed by the monomer (styrene (St), 1.0 g, 9.6 mmol or methyl
methacrylate (MMA), 1.0 g, 10.0 mmol) and the initiator (ethyl-2-
bromo-isobutyrate, 7.5
l
L, 5.1 ꢂ 10ꢁ5 mol). The flask was immedi-
ately immersed in the oil bath at 110 °C and kept for a given time
(Tables 5 and 6). After heating was stopped, the reaction vessel was
cooled to room temperature and a large excess of MeOH was added
to precipitate the polymer formed. The resulting solids were fil-
tered, washed with methanol and redissolved in toluene, where
Dowex ion exchange resin was added to remove any metal com-
pound before analysis. The colorless solution was then decanted
and concentrated to a small volume (ꢃ1 mL). MeOH was added
again to precipitate the polymers, which were then dried in vacuo
for several hours. Polymers were characterized by SEC in CHCl3 and
by 1H NMR in CDCl3 at room temperature.
3. Results and discussion
3.1. Synthesis and characterization
The syntheses of ligands LxH3 (x = 3, 5, 8, 9) have been previ-
ously reported [31]. Ligands L10H3 and L13H3 (Fig. S1) have been
prepared by analogous protocols to those employed for installation
of acyl and aryl arms to the tripodal trisamino-amine (2-NH2-
C6H4)3N (TRPHEN), respectively. The synthesis of the CoII
complexes (1–6; Scheme 1 (minimal structure of first coordination

V. Bagchi et al. / Polyhedron 52 (2013) 78–90
83
Table 2
Selected interatomic distances (Å) for compounds 1–6.
1
2
3
4
5
6
Co(1)–N(1)
Co(1)–N(2)
Co(1)–N(3)
Co(1)–N(4)
2.171(2)
1.980(2)
2.035(2)
2.018(2)
2.070(2)
2.091(3)
1.934(4)
1.944(4)
1.947(4)
2.1359(19)
1.9799(19)
1.9895(19)
1.9911(17)
2.096(6)
1.994(6)
1.965(6)
1.941(5)
2.1833(18)
1.9644(18)
1.9569(18)
1.9555(18)
2.2270(13)
1.9930(13)
1.9884(14)
1.9669(13)
2.1092(14)
Co(1)–N(5)
Co(2)–N(5)
Co(2)–N(6)
Co(2)–N(7)
Co(2)–N(8)
2.103(6)
1.954(6)
1.964(6)
1.940(6)
0.223
2.1584(16)
1.9743(17)
1.9646(17)
1.9616(18)
0.311
Co(1)–[N(2), N(3), N(4)]a
Co(2)–[N(6), N(7), N(8)]a
0.238
0.173
0.295
0.319
0.190
0.277
a
Distance of Co from mean plane.
Table 3
Selected angles (o) for compounds 1–6.
1
2
3
4
5
6
N(2)–Co(1)–N(4)
N(2)–Co(1)–N(3)
N(4)–Co(1)–N(3)
N(2)–Co(1)–N(1)
N(4)–Co(1)–N(1)
N(3)–Co(1)–N(1)
N(5)–Co(1)–N(1)
N(2)–Co(1)–N(5)
N(4)–Co(1)–N(5)
N(3)–Co(1)–N(5)
N(8)–Co(2)–N(6)
N(8)–Co(2)–N(7)
N(6)–Co(2)–N(7)
N(8)–Co(2)–N(5)
N(6)–Co(2)–N(5)
N(7)–Co(2)–N(5)
123.54(9)
110.35(9)
118.66(16)
118.98(17)
119.97(16)
84.98(14)
85.31(15)
84.26(18)
112.64(8)
114.04(8)
126.74(8)
81.55(8)
80.63(8)
82.22(8)
114.4(2)
118.4(2)
123.4(2)
82.2(2)
84.5(2)
83.7(2)
115.74(8)
119.67(8)
117.18(8)
80.99(7)
81.21(7)
80.43(7)
119.08(5)
113.81(5)
119.47(5)
80.47(5)
80.36(5)
81.39(5)
178.05(5)
101.21(5)
99.58(5)
96.98(5)
117.74(9)
81.74(9)
79.77(8)
79.45(8)
175.64(9)
102.61(9)
97.91(9)
98.53(9)
123.3(2)
115.5(2)
118.4(3)
85.5(2)
84.2(2)
83.5(2)
116.02(7)
120.46(7)
117.67(7)
81.66(7)
82.16(7)
81.93(7)
Table 4
Electrochemical data for compounds 1–6.
Compound
Solvent
E1/2 (or Ep,a) (V vs. Fc+/Fc)
DE (mV)
[K(L3)CoII–NCMe]n (1)
[K(THF)6][(L5)CoII]ꢀ1.5THF (2)
[K(THF)2(L8)CoII]n (3)
MeCN
DMF
DMF
DMF
DMA
MeCN
ꢁ0.090
72
102
ꢁ0.258
0.719
[K2(L9)2CoII
]
(4)
ꢁ0.654, ꢁ0.500
88, 75
78
2
n
{[K2(DMA)3(L10)2CoII2]ꢀ0.5Et2O}n (5)
0.559
[K(NCMe)3(L13)CoII–NCMe] (6)
ꢁ0.665
Table 5
Reactions of compounds 1–6 with styrene (St).a
Entry
Compound
t (h)
Yield (%)
Mn (th)b
Mn (exp)c
Mw/Mn
1
2
3
4
5
6
7
8
9
[K(L3)CoII–NCMe]n (1)
16
20
20
2
18
22
2
16
22
16
15
14
3000
2800
1200
d
d
[K(THF)6][(L5)CoII]ꢀ1.5THF (2)
9600
7100
2.78
2.39
1.47
1.56
1.57
1.82
2.00
1.97
1.54
12e
12
[K(THF)2(L8)CoII]n (3)
2400
81900
90f
86
18000
17200
10200
17200
3200
110500
57400
86100
38900
5700
[K2(L9)2CoII
] (4)
2 n
{[K2(DMA)3(L10)2CoII2]ꢀ0.5Et2O}n (5)
51
86f
16
[K(NCMe)3(L13)CoII–NCMe] (6)
–
10
21
4200
326500
a
b
c
d
e
f
Conditions: catalyst (0.01 mmol), styrene (9.6 mmol), ethyl-2-bromo-isobutyrate (0.05 mmol), 1.0 mL toluene, reflux at 110 °C.
Mn (th) = (g of monomer/mol initiator) ꢂ % yield.
By SEC in CHCl3 at r.t.
Very broad molecular weight distribution.
Ethyl-2-bromo-isobutyrate (0.10 mmol).
2.0 mL of toluene.

84
V. Bagchi et al. / Polyhedron 52 (2013) 78–90
Table 6
Reactions of compounds 1–6 with methyl methacrylate (MMA).a
Entry
Compound
t (h)
Yield (%)
Mn (th)b
Mn (exp)c
Mw/Mn
1
2
3
4
5
6
7
8
9
10
11
12
13
[K(L3)CoII–NCMe]n (1)
20
16
15
15
21
16
15
17
15
15
15
21
17
5
3d
30
33e
13f
6
1000
600
6000
3300
870
1200
2400
14000
14400
6000
8000
9000
1800
14,500
16700
37900
24500
20700
439800
300800
51400
67000
57300
55800
54700
356700
1.26
1.16
1.43
1.40
1.26
1.24
1.29
1.52
2.00
1.91
1.54
1.58
1.20
[K(THF)6][(L5)CoII]ꢀ1.5THF (2)
[K(THF)2(L8)CoII]n (3)
12d
70
72
60e
40
45d
9
[K2(L9)2CoII
] (4)
2 n
{[K2(DMA)3(L10)2CoII2]ꢀ0.5Et2O}n (5)
[K(NCMe)3(L13)CoII–NCMe] (6)
–
a
b
c
d
e
f
Conditions: catalyst (0.01 mmol), methyl methacrylate (10 mmol), ethyl-2-bromo-isobutyrate (0.05 mmol), 1.0 mL toluene, reflux at 110 °C.
Mn (th) = (g of monomer/mol initiator) ꢂ % yield.
By SEC in CHCl3 at r.t.
2.0 mL of toluene.
Ethyl-2-bromo-isobutyrate (0.10 mmol).
Ethyl-2-bromo-isobutyrate (0.15 mmol).
[K(L3)CoII–NCMe] (1)
[K(THF)6][(L5)CoII]•1.5THF (2)
[K(THF)2(L8)CoII] (3)
n
n
[K2(L9)2CoII ] (4)
{[K2(DMA)3(L10)2CoII ]•0.5Et2O} (5) [K(NCMe)3(L13)CoII NCMe] (6)
2
n
2
n
Scheme 1.
sphere is shown)) has been accomplished by (i) deprotonation of
the tripodal ligands (LxH3) with 3 equiv. of KH, and (ii) addition
of anhydrous CoCl2 under inert atmosphere, followed by filtration
to separate the insoluble KCl. The reactions are carried out in THF,
except for the reaction with L10H3, in which case DMA is used, be-
cause of the low solubility of the ligand in THF. For the isolation of
the crude CoII complexes, the filtrate is concentrated and layered
with pentane over THF, or diethylether over DMA. Crystals suitable
for X-ray diffraction analysis are obtained at low temperature from
concentrated THF solutions (3, 4), and from layering of two (THF/
pentane, 2) or three solvents (MeCN/Et2O/pentane, 1, 6; DMA/
Et2O/pentane, 5).
The solid-state structures of these compounds reveal similari-
ties in the coordination mode of the ligands, since in all cases the
four nitrogen-atom residues of the ligands are coordinated to the
metal center in a distorted trigonal-pyramidal geometry. However,
there are also significant differences. Two of the compounds are
five-coordinated (1 and 6), due to the presence of a good donor
solvent (MeCN) as a ligand occupying the fifth site. In spite of
the presence of THF or DMA, the remaining compounds are four-
coordinated (2–5). In sharp contrast, the corresponding
[K(THF)2][(L3)FeII–THF] and [K(DMA)][(L8)FeII–DMA)] species are
five-coordinated [31], although the L9 coordinated FeII and CoII
(4) compounds are essentially isostructural, and do not include
any solvent molecules (coordinated or solvated). Another interest-
ing feature is the intricate role played by potassium ions in gener-
ating polymeric tertiary structures (compounds 1, 3–5) by
establishing contacts with ligand residues (largely Namido atoms,
amidato carbonyl moieties and aromatic units). A more detailed
analysis of the solid-state structures is provided below.
Among the aryl-armed CoII reagents (1, 2, 6; Figs. 1–3), the two
five-coordinate structures (1, 6) exhibit distorted trigonal bipyra-
midal geometries, with that of
6 being the least distorted
(s
= 0.97 (6), 0.87 (1)). Compound 6 also possesses the most rigid

V. Bagchi et al. / Polyhedron 52 (2013) 78–90
85
Fig. 1. ORTEP drawing of [K(L3)CoII–NCMe]n (1) with 50% probability displacement
ellipsoids. Hydrogen atoms are omitted for clarity.
Fig. 3. ORTEP drawing of [K(NCMe)3(L13)CoII–NCMe] (6) with 50% probability
displacement ellipsoids. Hydrogen atoms are omitted for clarity.
Fig. 2. ORTEP drawing of the anion of [K(THF)6][(L5)CoII]ꢀ1.5THF (2) with 50%
probability displacement ellipsoids. Hydrogen atoms are omitted for clarity.
Fig. 4. ORTEP drawing of [K(THF)2(L8)CoII]n (3) with 50% probability displacement
ellipsoids. Hydrogen atoms are omitted for clarity.
cavity among the three aryl-bearing reagents, due to the 2,6-
rather than 3,5-aryl substitution, which locks the aryl arms in a
perpendicular position versus the corresponding phenylene groups
of the TRPHEN framework. The four-coordinate species 2 is a
benchmark case, because this is the only CoII compound reported
here, in which the K+ ion is solely coordinated by solvent molecules
(THF), leaving the [(L5)CoII]ꢁ anion structure unperturbed by any
potassium close contacts. As a result the average CoII–Namido bond
distance in 2 is the shortest available (1.942(7) Å) and with the
smallest deviation, not only with respect to the five-coordinate 1
and 6, but also among all four-coordinate species. Similarly, com-
pound 2 has the shortest CoII displacement (0.173 Å) from the
mean plane generated by the three equatorial Namido atoms. In con-
trast, compound 1 shows significant elongation of the Co–N(3) and
Co–N(4) bonds due to the interference of the K+ ion, which estab-
lishes close contacts between these two Namido moieties in adjacent
molecules, thus giving rise to an one-dimensional polymeric struc-
ture. In addition, K+–arene interactions, involving phenylene moie-
ties of the TRPHEN framework, are evident in 1 and 6. In the latter
case, these are the only contacts between the K+ ion and the ligand
(phenylene rings between atoms N(1) and N(2) and, to a lesser
extent, N(1) and N(3)), resulting in modest elongation of the
corresponding Co–N(2) and Co–N(3) bonds. Otherwise, 6 is a
molecular entity because the coordination of K+ is limited to K+–
arene interactions with the same ligand and solvation by three
MeCN molecules.
The acyl-armed CoII reagents 3 and 5 (Figs. 4 and 5) are strictly
four-coordinate species, in keeping with cobalt’s late transition-
metal character, and are thus devoid of the complications observed
in analogous FeII reagents, in which the acyl group contributes to
metal coordination via amidato –N–C(R)@Oꢀ ꢀ ꢀM or –N@C(R)–O–
M modes. As a matter of fact, the carbonyl moieties are oriented
exo with respect to the cavity of the vacant coordination site.
The strong electron-withdrawing character of the CF3CO– arm in
3 is evident in the longer average CoII–Namido bond distance
(1.987(6) Å) with respect to that of 5 (1.963(7) Å). Both structures
demonstrate complicated three-dimensional architectures, owing
to solvated K+ ions connecting acyl derived carbonyl moieties.
The tertiary structure of 3 is simpler and reminiscent of the corre-
sponding FeII-containing structure, inasmuch as equivalent
[K(THF)2]+ ions are coordinated by all three carbonyl residues

86
V. Bagchi et al. / Polyhedron 52 (2013) 78–90
Fig. 5. ORTEP drawing of {[K2(DMA)3(L10)2CoII2]ꢀ0.5Et2O}n (5) with 50% probability displacement ellipsoids. Hydrogen atoms are omitted for clarity.
Fig. 6. ORTEP drawing of [K2(L9)2CoII
] (4) with 50% probability displacement ellipsoids. Hydrogen atoms are omitted for clarity.
2 n
(O(1), O(2), O(3)), each oxygen belonging to a different molecule.
The corresponding structure of 5 distinguishes between two inde-
pendent CoII sites (Co(1) and Co(2)), differing slightly in metrical
parameters, and two solvated K+ ions ([K(DMA)]+ for K(1) and
[K(DMA)2]+ for K(2)), each interconnecting three CoII sites. K(1) is
associated with carbonyl residues (O(3), O(5)) of two separate
Co(1) sites and a carbonyl residue (O(7)) of one Co(2) site, whereas
K(2) is coordinated by carbonyl residues (O(9), O(11)) from two
different Co(2) sites and a carbonyl residue (O(1)) belonging to a
Co(1) site. Therefore, all carbonyl moieties associated with Co(1)
and Co(2) sites are involved in close contacts with K+ ions.
due course, as part of a study that includes a wider family of li-
gands, to further advance our understanding of solution structure.
In the meantime, the electrochemical data reported below in a
variety of solvents, support the notion that, in many instances,
well-defined mono-anionic structures are present in solution to
support reversible one-electron transfer chemistry.
3.2. Cyclic voltammetry
The electrochemical properties of compounds 1–6 have been
studied by cyclic voltammetry. Electrochemical data for the CoII/
CoIII couple are listed in Table 4 and the corresponding waves are
shown in Fig. 7. The complete voltammograms are shown in
Figs. S2–S7. Compound [K(NCMe)3][(L13)CoII–NCMe] (6), which
bears the electron-donating aryl substituent 2,6-(CH3)2 is oxidized
at the most negative potential. The presence of the electron-with-
drawing aryl substituents 3,5-(CF3)2 and 3,5-Cl2 in [K(L3)CoII–
NCMe]n (1) and [K(THF)6][(L5)CoII]ꢀ1.5THF (2) respectively, shifts
the oxidation redox potential to more positive values. Complex [K2-
(L9)2CoII2]n (4) shows two overlapping redox waves, which may re-
flect the two distinct CoII ions observed in the repeating dinuclear
unit of the solid-state structure. The potential values are close to
that of compound 6 and are in agreement with the electron-rich
nature of the arylalkylamido-amine ligand. Finally, as expected,
the electron-withdrawing COOtBu and COCF3 substituents of {[K2-
(DMA)3(L10)2CoII2]ꢀ0.5Et2O}n (5) and [K(THF)2(L8)CoII]n (3) cause
significant anodic shift. The assignment of these irreversible waves
Finally, compound 4 (Fig. 6) is the sole example of an alkyl-armed
ligand, and, as in the case of the corresponding FeII species [31],
exhibits a unique geometry, featuring a repeating –[Co(1)–K(1)–
Co(2)–K(2)]n–sequence. TheK(1)ioninterconnectsresiduesbelong-
ing to ligands attached to Co(1) and Co(2), namely atoms N(2) and
N(6), and phenylene rings via K+–( x-arene) contacts (phenylenes
g
between N(1)/N(4) (x = 6) and N(5)/N(7) (x = 3)). Ion K(2) is largely
involved in K+–( 6-arene) interactions (phenylene groups between
g
N(1)/N(2) and N(5)/N(8)) and long-range contacts with N(3) and
N(7). As a result, all Co–Namido bond distances are affected to various
degrees, the most elongated one being that of Co(1)–N(2).
As has been the case with the corresponding FeII compounds, it
is exceedingly difficult to determine whether the structural com-
plexity observed in the solid state, largely due to intimate K+ con-
tacts with ligand and solvent moieties, is retained intact in fluid
matrices. Detailed EPR and magnetic data will be presented in

V. Bagchi et al. / Polyhedron 52 (2013) 78–90
87
act for a given time. In some cases, high-viscosity mixtures had
formed at that time. The catalyst/monomer/initiator ratio was 1/
1000/5, unless otherwise stated. The polymers, polystyrene (PS)
and poly(methyl methacrylate) (PMMA), were obtained after addi-
tion of MeOH and were purified by Dowex ion exchange resin in
toluene solutions.
40
30
20
10
0
1
2
3
4
5
6
From the data shown in Table 5, higher yields of PS have been
obtained with compounds 3–5. The reaction mixtures containing
3 and 5 had gelled after 2 h (entries 4 and 7). For this reason, they
were repeated with 2.0 mL of toluene instead of 1.0 mL. It is nota-
ble however that the reaction with 5 gave high yields (51%) in a
short period of time. When a larger amount of solvent (2.0 mL; en-
try 8) was used, the reaction proceeds in a more controlled fashion,
giving a polymer with molecular weight closer to the theoretical
value, although with a bimodal distribution (Fig. S8), which causes
broadening of the molecular weight distribution (Mw/Mn = 2.00).
Among the three systems, 4 and 5 showed better control over
the molecular weight of the PS obtained (entries 6 and 8), while
compounds 3 and 4 gave PS with bimodal distribution and yet rel-
atively narrow (1.56 and 1.57, respectively; entries 5 and 6;
Figs. S9 and S10). The other three compounds show low reactivity
(entries 1–3 and 9), yielding polymers with broad molecular
weight distributions.
Similar reactivity has been observed in MMA polymerization in
most cases, with compounds 4 and 5 being the most reactive in
terms of polymer yields. The only notable exception is compound
3 (Table 6, entries 6 and 7), which is the most reactive compound
for St polymerization, and one of the least reactive for MMA poly-
merization. In fact, the complex has no control over the reaction, as
the molecular characteristics of the polymers formed do not differ
from the blank experiment (entry 13). This result is not unex-
pected, since compound 3 is totally insoluble in the reaction mix-
ture. The molecular weight of PMMA obtained with compound 4
shows the best agreement with the theoretical value, compared
to the other samples, and the molecular weight distribution is
not very broad (Mw/Mn = 1.52; entry 9; Fig. S11). Compound 5 gives
high yields of PMMA, although with higher molecular weights and
i
-10
-1.2
-0.8
-0.4
0.0
0.4
0.8
E (V vs. Fc+/Fc)
Fig. 7. Cyclic voltammograms (CoII/CoIII redox couple) of compounds [K(L3)CoII–
NCMe]n (1) and [K(NCMe)3(L13)CoII–NCMe] (6) in MeCN/[(n-Bu)4N]PF6, [K(THF)6]-
[(L5)CoII]ꢀ1.5THF (2), [K(THF)2(L8)CoII]n (3) and [K2(L9)2CoII
]
(4) in DMF/
2
n
[(n-Bu)4N]PF6, and {[K2(DMA)3(L10)2CoII2]ꢀ0.5Et2O}n (5) in DMA/[(n-Bu)4N]PF6, as
indicated, with a Au disk electrode (1.6 mm in diameter); scan rate 0.1 V/s.
to the CoII/CoIII couple is tentative. Contributions from ligand-cen-
tered events cannot be excluded at the present time, especially for
compound 3, for which a large ip,a value is observed. These results
are consistent with the electronic character of the ligands and prior
electrochemical data obtained for the analogous FeII complexes
[31], although the potentials of the CoII compounds have been
shifted anodically, as expected, by a factor of 0.4–0.7 V.
3.3. Polymerization results
Compounds 1, 2, 4, and 6 have properties that make them
promising reagents for use as ATRP catalysts. They provide revers-
ible or semi-reversible CoII/CoIII couples at easily accessible poten-
tials, as shown by the electrochemical data. They have either a
vacant coordination site, or a weakly bound ligand (solvent mole-
cule), which makes it possible for a halide or (pseudohalide) ligand
to enter the coordination sphere. Compounds 3 and 5, which fea-
ture an irreversible CoII/CoIII couple, have also been studied, as
the halidophilicity is in some cases a factor of equal or even higher
importance. Indeed, it has been found that the activity of an ATRP
catalyst mainly depends on the redox potential of the Mn+/M(n+1)+
couple and the halidophilicity of the M(n+1)+ complex. For Cu-based
catalysts the key factor seems to be the redox potential, for Ru- and
Os-complexes the halidophilicity, while for other metal catalysts
the two factors have intermediate effects [42]. In addition, since
ATRP is not the only possible pathway for Co-induced radical poly-
merizations, other parameters, such as the strength and reactivity
of CoIII–C or CoIII–H bonds, can play a dominant role.
The catalytic activity of compounds 1–6 towards the polymeri-
zation of styrene (St) and methyl methacrylate (MMA) has been
studied in toluene solutions at 110 °C using ethyl-2-bromo-isobu-
tyrate as initiator. The results are summarized in Tables 5 and 6.
Blank experiments are also included for comparison purposes.
Compounds 1, 5 and 6 are readily soluble in toluene, 2 is partially
dissolved, whereas 3 and 4 are insoluble. However, the solutions
become homogeneous after the addition of the monomer and the
initiator, with the only exception of compound 3, which gets partly
dissolved during the reaction with St and remains insoluble during
the reaction with MMA. The reaction mixtures were allowed to re-
bimodal molecular weight distributions (entries 9 and 10;
Fig. S12). With compounds 2 and 6 lower yields are obtained (en-
tries 3, 11 and 12), but the metal complex seems to have a better
control over the reaction, giving PMMA with relatively narrow
molecular weight distribution, especially with compound
2
(1.40). Addition of 0.10 and 0.15 mmol of initiator (entries 4 and
5) does not significantly alter the controllability of the reaction,
although the GPC plots have different characteristics (Fig. 8), which
will be discussed below. In the latter case, the yield is reduced. Fi-
nally, 1 is the least reactive of the compounds studied (entries 1
and 2).
A common characteristic of all experiments is that the experi-
mental molecular weights are higher than the theoretical ones,
which shows that termination reactions take place during the
polymerization. Another interesting point is that in many cases a
bimodal distribution is observed (PS and PMMA obtained with
compounds 2–5), resulting in the broadening of the molecular
weight distribution. This is indicative of the presence of at least
two reactive centers, and given that the molecular weight distribu-
tions are not very broad (1.40–2.00, except for the 2/St system),
these results strongly indicate that two parallel mechanisms are
in operation. The reactions of compound 2 with MMA further sup-
port this argument. Three experiments have been carried out, with
different quantities of initiator (0.05, 0.10 and 0.15 mmol). In all
cases, the GPC plots show bimodal distributions, but the ratio of
the two peaks differ from one experiment to the other, as shown
in Fig. 8. These results show that the reaction proceeds with two
distinct mechanisms, and that the amount of initiator used can fa-
vor one over the other.

88
V. Bagchi et al. / Polyhedron 52 (2013) 78–90
Fig. 9. SEC traces for PS obtained from the reaction of [K2(L9)2CoII
]
(4)
2
n
(0.01 mmol) with St (9.6 mmol) and ethyl-2-bromo-isobutyrate (0.05 mmol) in
toluene (2.0 mL) under reflux at 110 °C (top) and PS obtained after the second
addition of St (9.6 mmol, bottom).
Fig. 8. SEC traces for PMMA obtained from the reaction of [K(THF)6][(L5)CoII]-
ꢀ1.5THF (2) (0.01 mmol) with MMA (10.0 mmol) and ethyl-2-bromo-isobutyrate
(0.05 mmol, top; 0.10 mmol, middle; 0.15 mmol, bottom) in toluene (1.0 mL) under
reflux at 110 °C.
In order to gain more evidence on the controlled nature of the
polymerization and the operation of the ATRP mechanism, we
examined the ability of the polymers formed to reinitiate the poly-
merization reaction.
In the first experiment, St has been polymerized with com-
pound [K2(L9)2CoII
] (4) for 20 h. A small aliquot of the solution
2 n
was taken and MeOH was added, yielding PS with Mn = 50200
and Mw/Mn = 1.80 (Fig. 9). To the rest of the reaction mixture, St
was added and was left to react for another 20 h, providing PS with
Mn = 67000 and Mw/Mn = 1.59 (Fig. 9). The fact that the molecular
weight has increased and the molecular weight distribution has
decreased is a strong indication that the polymerization does not
proceed through free radicals but in a rather controlled way, where
the CoII complex plays a crucial role.
For the second experiment, MMA was polymerized with
[K(THF)6][(L5)CoII]ꢀ1.5THF (2) and 0.05 mmol of the initiator, be-
cause under these conditions one mechanism seems to be domi-
nant. After the end of the reaction (20 h), the PMMA formed was
precipitated and characterized (Mn = 23700 and Mw/Mn = 1.33).
Then, it was redissolved in toluene, where the metal complex
and St were added, and kept under reflux for 20 h) yielding a
PMMA-b-PS block copolymer with Mn = 72400 and Mw/Mn = 1.60.
From the comparison of the chromatographic traces (Fig. 10), it
is obvious that the peak of PMMA at 36.9 min is absent in the
GPC of the copolymer, which indicates that the end-Br groups of
the PMMA chains were exclusively employed as initiation sites
for the polymerization of St or in other words that PMMA served
as macroinitiator for the synthesis of PMMA-b-PS block copoly-
mers. The bimodality of the GPC trace of the final product can be
attributed to possible termination reactions and/or the increase
of viscosity of the reaction medium leading to polymerization un-
der non-homogeneous conditions. The successful synthesis of the
block copolymer was verified by 1H NMR spectroscopy (Fig. S14;
PMMA-b-PS 20–80% mol). This result confirms that the polymeri-
zation does not proceed through a coordination mechanism of
the monomer to the metal center, since in this case the results
would be a mixture of PMMA and PS homopolymers.
Fig. 10. SEC traces for PS obtained from the reaction of [K(THF)6][(L5)CoII]ꢀ1.5THF
(2) (0.01 mmol) with MMA (10.0 mmol) and ethyl-2-bromo-isobutyrate
(0.05 mmol) in toluene (2.0 mL) under reflux at 110 °C (top) and PMMA-b-PS
obtained after the reaction of
2 (0.01 mmol) with St (9.6 mmol) and PMMA
(0.05 mmol) in toluene (2.0 mL) under reflux at 110 °C (bottom).
results are summarized in Table 7 and a representative spectrum
in shown in Fig. S13. The polymers obtained by conventional rad-
ical polymerization are mainly syndiotactic (rr:rm:mm = 64:32:4),
due to the repulsion between the substituents of the chain end and
less-controlled chain propagation [43]. The sequence distribution
of the tacticity of PMMA is approximately Bernoullian. Similar
microstructural results have been reported from ATRP of MMA
using various catalytic species, such as those based on Rh [44], Fe
[45,46], Ni [47,48], Cu [49] and Ru [50] complexes. The polymers
obtained in this study were predominately syndiotactic, consistent
with radical polymerization. Compounds 1 and 2 exhibit unusually
high ratios of isotactic triads. It is known that several CoII com-
plexes act as efficient chain transfer agents during methacrylate
polymerization. The CoII radical may interact with the propagating
radical and induce single-handed helical structures in the polymer-
ization of bulky methacrylates [51]. The polymerization proceeds
with a different mechanism from typical catalytic chain transfer
although the detailed mechanism is not clear [52]. It is possible
The tacticity of the PMMA macromolecules, formed in the pres-
ent study, has been determined by 1H NMR spectroscopy. The

V. Bagchi et al. / Polyhedron 52 (2013) 78–90
89
Table 7
exhibiting anodic waves at high positive voltages, for which
ligand participation cannot be excluded.
Tacticity of poly(methyl methacrylate) (PMMA).a,b
(4) Our preliminary results show that compounds 2–6 can
induce the controlled radical polymerization (CRP) of sty-
rene and methyl methacrylate. Some compounds (3, 4 and
5) provide good yields (60–90%; 3 is active only in St poly-
merization), whereas others (2 and 6) have better control
over the polymerization reactions, albeit with lower yields
(12–16% for PS and 30–45% for PMMA). Although a firm cor-
relation between the reactivity and the structure or the
redox potential cannot be made at this point, the steric fac-
tor seems to play an essential role, since compounds that
feature a less hindered fifth coordination site show the high-
est reactivity and better control over the polymerization
reaction.
(5) The polymers obtained are predominately syndiotactic,
consistent with radical polymerization, with two exceptions
that exhibit unusually high ratios of isotactic triads. In
many cases bimodal distributions are observed, although
the molecular weight distributions are not very broad
(1.40–2.00), and this strongly indicates that two parallel
mechanisms are in operation. Among the three predominant
mechanisms (ATRP, OMRP and CCTP) that have been estab-
lished for cobalt-mediated CRP reactions, the first two seem
to be more likely for our systems. A detailed mechanistic
study is under way.
Entry
Compound
mm (%)
mr (%)
rr (%)
1
2
3
4
5
6
7
8
9
10
11
12
13
[K(L3)CoII–NCMe]n (1)
28
8
16
10
8
7
5
10
6
10
7
31
36
33
37
39
38
35
38
40
39
39
40
38
41
56
51
53
53
55
60
52
54
51
54
53
56
[K(THF)6][(L5)CoII]ꢀ1.5THF (2)
[K(THF)2(L8)CoII]n (3)
[K2(L9)2CoII
]
n
(4)
2
{[K2(DMA)3(L10)2CoII2]ꢀ0.5Et2O}n (5)
[K(NCMe)3(L13)CoII–NCMe] (6)
7
6
–
Calculated from 1H NMR spectra.
The experiments are presented in the same order as in Table 6.
a
b
that a similar situation exists, at least to some extent, with com-
pounds 1 and 2 leading to increased isotacticities. Further mecha-
nistic studies are needed to clarify this finding.
Although a firm correlation between the reactivity and the
structure or the redox potential of the complexes cannot be made
at this time, our preliminary results indicate that the steric rather
than the electronic factor is the most important in the polymeriza-
tion reactions, since compounds 4 and 5 that show the highest
reactivity feature a less hindered fifth coordination site. The exper-
imental data, in most cases, suggest that more than one mecha-
nism is operative. Between the three mechanisms that have been
established for cobalt-catalyzed CRP reactions, namely, ATRP,
OMRP and CCTP, the last one seems to be the least possible in
our case, because the lack of extra coordination sites in the metal
compounds under study would make the b-H elimination very
challenging. Finally, in the unique case of compound 3, an hetero-
geneous mechanism for St polymerization cannot be excluded.
The experimental findings strongly support the conclusion that
the CoII complexes prepared in this work, were efficiently involved
in the controlled radical polymerization of MMA and St. However,
more work is needed to identify the experimental polymerization
conditions that will provide the best control over the structural
and molecular characteristics.
Acknowledgments
We gratefully acknowledge financial support by the NSF
(CHE-0412959) (to P.S.) and by the Special Account of Research
Grants of the University of Athens.
Appendix A. Supplementary data
CCDC 879541, 879542, 879543, 879544, 879545, 879546 and
879547 contain the supplementary crystallographic data for 1–6
and L13H3, respectively. These data can be obtained free of
charge via http://www.ccdc.cam.ac.uk/conts/retrieving.html, or
from the Cambridge Crystallographic Data Centre, 12 Union Road,
Cambridge CB2 1EZ, UK; fax: (+44) 1223-336-033; or e-mail:
deposit@ccdc.cam.ac.uk. Supplementary data associated with this
article can be found, in the online version, at http://dx.doi.org/
10.1016/j.poly.2012.11.020.
4. Conclusions
The following are the main results and conclusions of the pres-
ent research:
(1) A series of CoII reagents, coordinated by a family of tripodal
[N3N]3ꢁ ligands bearing a triphenylamido-amine core and
aryl, acyl and alkyl arms, have been synthesized and charac-
terized by a variety of physico-chemical techniques.
References
[1] A. Debuigne, R. Poli, C. Jérôme, R. Jérôme, C. Detrembleur, Prog. Polym. Sci. 34
(2009) 211.
[2] B.B. Wayland, G. Poszmik, S.L. Mukerjee, J. Am. Chem. Soc. 116 (1994) 7943.
[3] L.D. Arvanitopoulos, M.P. Greuel, H.J. Harwood, Polym. Prepr. (Am. Chem. Soc.,
Div. Polym. Chem.) 35 (1994) 549.
[4] T. Pintauer, K. Matyjaszewski, Chem. Soc. Rev. 37 (2008) 1087.
[5] M.A. Fernández-Zúmel, K. Thommes, G. Kiefer, A. Sienkiewicz, K. Pierzchala, K.
Severin, Chem. Eur. J. 15 (2009) 11601.
[6] C. Ricardo, T. Pintauer, Chem. Commun. (2009) 3029.
[7] W. Tang, Y. Kwak, W. Braunecker, N.V. Tsarevsky, M.L. Coote, K. Matyjaszewski,
J. Am. Chem. Soc. 130 (2008) 10702.
[8] X. Luo, X. Zhao, S. Xu, B. Wang, Polymer 50 (2009) 796.
[9] Z.-X. Huang, Y.-M. Zhang, H. Li, Y.-G. Liu, Macromol. Chem. Phys. 209 (2008)
825.
[10] M.-S. Weiser, R. Mülhaupt, J. Polym. Sci., Part A: Polym. Chem. 43 (2005) 3804.
[11] Z. Zhang, W. Zhang, X. Zhu, Z. Cheng, J. Zhu, J. Polym. Sci., Part A: Polym. Chem.
45 (2007) 5722.
[12] Z.H. Li, Y.M. Zhang, M.Z. Xue, L. Zhou, Y.G. Liu, J. Polym. Sci., Part A: Polym.
Chem. 43 (2005) 5207.
[13] Z. Huang, Y. Zhang, H. Li, Y. Liu, Appl. Catal., A 332 (2007) 192.
[14] Z.X. Huang, Y.M. Zhang, H. Li, Y.H. Luan, Y.G. Liu, J. Polym. Sci., Part A: Polym.
Chem. 46 (2008) 1416.
(2) X-ray structural analysis reveals that all CoII complexes
demonstrate a distorted trigonal pyramidal geometry gener-
ated by three ligand-derived Namido residues and an apical
Namine atom. In the presence of a good donor solvent such
as MeCN, five-coordinate trigonal bipyramidal geometries
are obtained, otherwise four-coordinate species are pre-
ferred in contrast to the corresponding FeII reagents. Owing
to K+ contacts with aromatic, Namido and/or Oacyl residues
belonging to adjacent molecules, polymeric structures are
observed in four compounds.
(3) Electrochemical data establish that all aryl- and alkyl-armed
ligands give rise to reversible or semi-reversible CoII/CoIII
couples, with redox potentials ranging by as much as
575 mV. On the other hand, the electron-withdrawing
acyl-armed ligands demonstrate irreversible features,

90
V. Bagchi et al. / Polyhedron 52 (2013) 78–90
[15] K. Matsubara, M. Matsumoto, J. Polym. Sci., Part A: Polym. Chem. 44 (2006)
4222.
[16] H. Kaneyoshi, K. Matyjaszewski, J. Polym. Sci., Part A: Polym. Chem. 45 (2007)
447.
[33] M.B. Jones, C.E. MacBeth, Inorg. Chem. 46 (2007) 8117.
[34] L.C. Chu, K.I. Hardcastle, C.E. MacBeth, Inorg. Chem. 49 (2010) 7521.
[35] M.B. Jones, B.S. Newell, W.A. Hoffert, K.I. Hardcastle, M.P. Shores, C.E. MacBeth,
Dalton Trans. 39 (2010) 401.
[17] A. Debuigne, J.-R. Caille, C. Detrembleur, R. Jérôme, Angew. Chem., Int. Ed. 44
(2005) 3439.
[36] M.B. Jones, K.I. Hardcastle, C.E. MacBeth, Polyhedron 29 (2010) 116.
[37] M.B. Jones, K.I. Hardcastle, K.S. Hagen, C.E. MacBeth, Inorg. Chem. 50 (2011)
6402.
[38] S.K. Sharma, P.S. May, M.B. Jones, S. Lense, K.I. Hardcastle, C.E. MacBeth, Chem.
Commun. 47 (2011) 1827.
[39] Bruker, SMART, Bruker AXS Inc., Madison, Wisconsin, USA, 2002.
[40] Bruker, SAINT and SADABS, Bruker AXS Inc., Madison, Wisconsin, USA, 2008.
[41] G.M. Sheldrick, Acta Crystallogr., Sect. A 64 (2008) 112.
[42] F. di Lena, K. Matyjaszewski, Prog. Polym. Sci. 35 (2010) 959.
[43] K. Matyjaszewski, T.P. Davis (Eds.), Handbook of Radical Polymerization,
Wiley–Interscience, 2002 (Chapter 13).
[44] G. Moineau, C. Granel, P. Dubois, R. Jérôme, P. Teyssié, Macromolecules 31
(1998) 542.
[45] T. Ando, M. Kamigaito, M. Sawamoto, Macromolecules 30 (1997) 4507.
[46] K. Matyjaszewski, M. Wei, J. Xia, N.E. McDermott, Macromolecules 30 (1997)
8161.
[18] A. Debuigne, J.-R. Caille, N. Willet, R. Jérôme, Macromolecules 38 (2005) 9488.
[19] H. Kaneyoshi, K. Matyjaszewski, Macromolecules 39 (2006) 2757.
[20] A. Debuigne, J.-R. Caille, R. Jérôme, Angew. Chem., Int. Ed. 44 (2005) 1101.
[21] B.B. Wayland, L. Basickes, S. Mukerjee, M. Wei, M. Fryd, Macromolecules 30
(1997) 8109.
[22] Z. Lu, M. Fryd, B.B. Wayland, Macromolecules 37 (2004) 2686.
[23] S.C.J. Pierik, A.M. van Herk, Macromol. Chem. Phys. 204 (2003) 1406.
[24] A.A. Gridnev, S.D. Ittel, Chem. Rev. 101 (2001) 3611.
[25] D.M. Haddleton, E. Depaquis, E.J. Kelly, D. Kukulj, S.R. Morsley, S.A.F. Bon, M.D.
Eason, A.G. Steward, J. Polym. Sci., Part A: Polym. Chem. 39 (2001) 2378.
[26] D. Kukulj, J.P.A. Heuts, T.P. Davis, Macromolecules 31 (1998) 6034.
[27] J.P.A. Heuts, D. Kukulj, D.J. Forster, T.P. Davis, Macromolecules 31 (1998) 2894.
[28] D. Kukulj, T.P. Davis, K.G. Suddaby, D.M. Haddleton, R.G. Gilbert, J. Polym. Sci.,
Part A: Polym. Chem. Ed. 35 (1997) 859.
[29] R. Çelenligil-Çetin, P. Paraskevopoulou, R. Dinda, R.J. Staples, E. Sinn, N.P. Rath,
P. Stavropoulos, Inorg. Chem. 47 (2008) 1165.
[47] G. Moineau, M. Minet, P. Dubois, P. Teyssié, T. Senninger, R. Jérôme,
Macromolecules 32 (1999) 27.
[30] R. Çelenligil-Çetin, P. Paraskevopoulou, N. Lalioti, Y. Sanakis, R.J. Staples, N.P.
Rath, P. Stavropoulos, Inorg. Chem. 47 (2008) 10998.
[31] P. Paraskevopoulou, A. Lin, Q. Wang, D. Pinnapareddy, R. Acharrya, R. Dinda, R.
Çelenligil-Çetin, G. Floros, Y. Sanakis, A. Choudhury, N.P. Rath, P. Stavropoulos,
Inorg. Chem. 49 (2010) 108.
[48] C. Granel, P. Dubois, R. Jérôme, P. Teyssié, Macromolecules 29 (1996) 8576.
[49] D.M. Haddleton, C.B. Jasieczek, M.J. Hannon, A.J. Shooter, Macromolecules 30
(1997) 2190.
[50] M. Kato, M. Kamigaito, M. Sawamoto, T. Higashimura, Macromolecules 28
(1995) 1721.
[32] R. Çelenligil-Çetin, P. Paraskevopoulou, R. Dinda, N. Lalioti, Y. Sanakis, A.M.
Rawashdeh, R.J. Staples, E. Sinn, P. Stavropoulos, Eur. J. Inorg. Chem. (2008)
673.
[51] T. Nakano, Y. Okamoto, Macromolecules 32 (1999) 2391.
[52] A. Gridnev, J. Polym. Sci., Part A: Polym. Chem. Ed. 38 (2000) 1753.