Controlled radical polymerisation of methyl acrylate initiated by a well-defined cobalt alkyl complex

Article abstract of DOI:10.1039/b922030e

Cobalt-mediated radical polymerisation of methyl acrylate is effected at moderate temperature by both a square planar bis(β-ketoaminato)cobalt(ii) complex in the presence of V-70 diazo radical initiator, and by a well-defined cobalt(iii) alkyl derivative in the absence of a diazo complex. Polymerisation rates suggest that both reversible termination and degenerative transfer mechanisms are operative. The Royal Society of Chemistry.

Full text of DOI:10.1039/b922030e

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Controlled radical polymerisation of methyl acrylate initiated
by a well-defined cobalt alkyl complexw
Roger K. Sherwood,a Catherine L. Kent,a Brian O. Patrickb and W. Stephen McNeil*a
Received (in Cambridge, UK) 21st October 2009, Accepted 8th February 2010
First published as an Advance Article on the web 23rd February 2010
DOI: 10.1039/b922030e
Cobalt-mediated radical polymerisation of methyl acrylate is
effected at moderate temperature by both a square planar
bis(b-ketoaminato)cobalt(II) complex in the presence of V-70
diazo radical initiator, and by a well-defined cobalt(III) alkyl
derivative in the absence of a diazo complex. Polymerisation
rates suggest that both reversible termination and degenerative
transfer mechanisms are operative.
pathway can occur under higher radical concentrations in
which a radical species—either a growing polymer chain or a
radical injected by the initiator—displaces the trapped radical
bound to the metal (Scheme 1B).5,7,11
In either mechanism, the final products of such reactions are
cobalt-capped polymer chains, and while such species have
themselves been shown to act as effective initiators for
CMRP,3,4,8,12 examples of controlled radical polymerisations
using a well-defined cobalt(III) alkyl complex as the sole radical
source are exceedingly rare.13 We wish to report the controlled
radical polymerisation of methyl acrylate mediated by the
well-known (ketoaminato)cobalt(II) complex LCo (1)14
(L = bis(benzoylacetone)ethylenediaminate) using V-70 as a
radical initiator, and its comparison to the same polymerisation
effected by LCoEt (2) in the absence of V-70.
The controlled radical polymerization (CRP) of functionalized
alkenes is a powerful method to craft polymers with well-
controlled chain lengths and molecular weight distributions
via the addition of some mediating species that reduces the
concentration of growing radical chains.1 Among the various
methods of CRP, one garnering considerable attention is that
of cobalt-mediated radical polymerisation (CMRP),2 where
the mediating species is a cobalt complex that traps the
radicals by forming a Co–C bond. Various CMRP studies
have yielded systems capable of forming polymers and block
copolymers of acrylates or vinyl acetate (VOAc) in an
effectively living fashion.3–9 Such systems generally rely on
the use of diazo compounds as an initial source of radicals, the
transportation and low-temperature storage of which can be
problematic.10 Two competing mechanisms have been shown
to occur in CMRP reactions. Reversible termination (RT)
involves the trap and release of the radical chain via reversible
metal–carbon bond homolysis (Scheme 1A), and occurs when
the total radical concentration is less than that of the trapping
cobalt species. In contrast, a degenerative chain transfer (DT)
Compound 1z exhibits a solution magnetic moment of
1
2
1.99 mB, consistent with the low-spin S = configuration
expected for a square-planar CoII species. Our determination
of the solid-state structure of 1 conforms to those of previously
published15 structural reports establishing the square planar
geometry.
Compound 2 is best prepared by the treatment of 1 with an
excess of BEt3, which releases ethyl radicals upon exposure to
air.16y The alkyl radicals are trapped by 1 to afford 2 as an
air-stable diamagnetic complex. The structure of 2 is shown in
Fig. 1.z Unfortunately, the crystal exhibits a whole-molecule
disorder that limits high precision of the measured metrical
parameters (see ESIw for further details). However, the structure
unambiguously confirms the expected square pyramidal
geometry, comparable to that of similar square pyramidal
cobalt alkyl complexes.17 Upon addition of the alkyl group,
the Co–O and Co–N bond lengths increase slightly (on
average by B0.026 and B0.011 A, respectively), perhaps
contrary to the expectation associated with a formal oxidation
Scheme 1 CMRP mechanisms.
a Department of Chemistry, University of British Columbia Okanagan,
3333 University Way, Kelowna BC, Canada.
E-mail: s.mcneil@ubc.ca; Fax: +1 250 807 8005;
Tel: +1 250 807 8751
b Department of Chemistry, University of British Columbia,
2036 Main Mall, Vancouver BC, Canada
w Electronic supplementary information (ESI) available: CIF file for 2;
experimental data for X-ray structure determination of 2, synthetic
procedures and characterisation data for 1 and 2, experimental
procedures for polymerisation experiments (11 pages). CCDC [CCDC
NUMBER(S)]. For ESI and crystallographic data in CIF or other
electronic format see DOI: 10.1039/b922030e
Fig. 1 Molecular structure of 2, showing one of two disordered
orientations. Selected bond lengths [A] and angles [1]: Co–N1
1.867(9), Co–N2 1.861(9), Co–O1 1.865(8), Co–O2 1.867(8), Co–C23
1.991(6); O1–Co–O2 76.6(6), O1–Co–N1 98.6(14), N1–Co–N2
86.9(16), N2–Co–O2 97.1(12), O1–Co–C23 89.7(11), O2–Co–C23
99.3(7), N1–Co–C23 88.4(17), N2–Co–C23 96.5(8).
ꢀc
This journal is The Royal Society of Chemistry 2010
2456 | Chem. Commun., 2010, 46, 2456–2458

View Article Online
Fig. 4 Kinetic plot for polymerization of MA with 1 and V-70 in
toluene at 50 1C. [1]i : [V-70]i : [MA]i = 1 : 0.9 : 500, [1]i = 0.010 M.
Fig. 2 Variation of pMA mass (Mn) and polydispersity (Mw/Mn)
with monomer conversion in toluene at 50 1C. [1]i : [V-70]i : [MA]i =
1 : 0.6 : 500, [1]i = 0.010 M.
again suggesting a common RT mechanism independent of the
initial amount of V-70. After induction, the rate accelerates as
the concentration of radicals increases, and a DT mechanism
is presumed to apply.
from CoII to CoIII. However, the cobalt ion is low-spin in both
compounds, minimizing the effect on bond lengths of such a
change, and the small increase can be attributed primarily to
the raising of the cobalt atom in 2 B0.11 A above the mean
plane of the ketoaminate ligand donor atoms.
In both cases, the observed polydispersity gradually
increases over the course of the reaction, and the observed
polymer weights begin to drop relative to the theoretical value
(Fig. 2 and Fig. S1, see ESIw). These results suggest a gradual
loss of control of the polymerization. A likely cause is the
increasing role of non-reversible terminations, due in part to a
catalytic chain transfer mechanism,20 a commonly observed
effect at longer reaction times in CMRP, particularly at
somewhat elevated temperatures.2
Treatment of red-orange 1 with 0.6 equiv. V-70 and
500 equiv. methyl acrylate (MA) in toluene at 50 1C yields
poly(methyl acrylate) (pMA), a reaction accompanied by a
colour change to brown. As shown in Fig. 2, the polymer mass
is consistently somewhat above the theoretical value assuming
one growing polymer chain per cobalt centre, indicating that
not every molecule of 1 is involved in mediating the reaction.
Indeed, this must be the case: assuming an initiator efficiency
of 0.6,5 the total amount of radicals derived from V-70 never
equals that of 1. As well, the gradual decomposition of V-70
should result in initially low f values (f = Mn(theo)/Mn(obs))
that slowly increase as the reaction proceeds, as observed. The
plot of ln([MA]0/[MA]) vs. time (Fig. 3) illustrates a linear
first-order consumption of monomer for more than six V-70
half lives,18 indicating the maintenance of a constant radical
concentration during and beyond the decomposition of the
initiator.19 This suggests that the reaction occurs solely via
reversible release and trapping of the growing polymer chain
by 1 to form an organocobalt(III) intermediate (i.e. RT,
Scheme 1A). Increasing the concentration of V-70 to
0.9 equivalents yields the kinetic plot in Fig. 4. The calculated
induction period (until [Rꢁ]tot = [1]init) is 2.3 h, as observed.
As well, the induction period rate compares reasonably well to
that observed in Fig. 3 (0.060 ꢂ 0.010 hꢃ1 vs. 0.046 ꢂ 0.003 hꢃ1),
Reaction of brown 2 with 500 equiv. MA in toluene at 50 1C
initially affords pMA at much faster rates that follow non-
linear kinetics (Fig. 5). However, the reaction soon exhibits a
first-order loss of monomer comparable to that observed
in previous experiments during the RT induction period
(0.058 ꢂ 0.004 hꢃ1). This variation indicates that the generation
of alkyl radicals from ethyl complex 2 is slower than from a
polymer chain bound to 1, affording a lower initial concentration
of trapping agent 1.21 With less CoII initially available to act as
a trapping agent, fewer terminations result in faster polymer
growth (kobs
= 2 is
kp[MA](kd[CoIIIR]/kt[CoII]). After
consumed and the concentration of CoII is higher, new
radicals can only be generated by the RT equilibrium,
and the original rate is observed. Further evidence for this
conclusion is offered by MA polymerization in the presence of
a 1 : 1 mixture of 1 and 2. In this case, monomer loss exhibits
linear kinetics immediately at a rate identical to reactions
Fig. 3 Kinetic plot for polymerization of MA with 1 and V-70 in
toluene at 50 1C. [1]i : [V-70]i : [MA]i = 1 : 0.6 : 500, [1]i = 0.010 M.
Fig. 5 Kinetic plot for polymerization of MA with 2 in toluene at
50 1C. [2]i : [MA]i = 1 : 500, [2]i = 0.010 M.
ꢀc
This journal is The Royal Society of Chemistry 2010
Chem. Commun., 2010, 46, 2456–2458 | 2457

View Article Online
z Crystal data for 2: C24H27N2O2Co, M = 434.41, monoclinic,
a = 7.5525(2), b = 17.4630(6), c = 15.6446(5), b = 96.808(2)1,
U = 2048.8(1) A3, T = 173 K, space group P21/n, Z = 4, 31 367
reflections measured, 6288 unique (Rint = 0.061) and used in all
calculations, final wR(F2) = 0.139.
1 W. A. Braunecker and K. Matyjaszewski, Prog. Polym. Sci., 2007,
32, 93–146.
2 A. Debuigne, R. Poli, C. Jerome, R. Jerome and C. Detrembleur,
´ ´
Prog. Polym. Sci., 2009, 34, 211–239.
3 Z. Lu, M. Fryd and B. B. Wayland, Macromolecules, 2004, 37,
2686–2687.
´
4 A. Debuigne, J.-R. Caille and R. Jerome, Angew. Chem., Int. Ed.,
2005, 44, 1101–1104.
5 B. B. Wayland, C.-H. Peng, X. Fu, Z. Lu and M. Fyrd,
Macromolecules, 2006, 39, 8219–8222.
6 C.-H. Peng, M. Fyrd and B. B. Wayland, Macromolecules, 2007,
40, 6814–6819.
Fig. 6 Kinetic plot for polymerization of MA with 1 and 2 in toluene
at 50 1C. [1]i : [2]i : [MA]i = 1 : 500, [1]i = 0.010 M.
7 C.-H. Peng, J. Scricco, S. Li, M. Fyrd and B. B. Wayland,
Macromolecules, 2008, 41, 2368–2373.
8 A. Debuigne, Y. Champouret, R. Jerome, R. Poli and
´
C. Detrembleur, Chem.–Eur. J., 2008, 14, 4046–4059.
9 B. J. Langlotz, J. L. Fillot, J. H. Gross, H. Wadepohl and
L. H. Gade, Chem.–Eur. J., 2008, 14, 10267–10279.
using low concentrations of V-70 as initiator rather than 2
(Fig. 6, 0.046 ꢂ 0.003 h–1). As with V-70, using 2 as the
initiator yields polydispersities from 1.4 to 1.6 at longer
reactions times, the similar behaviour again indicative
of a common underlying mechanism once the initiator is
consumed. In other words, 1 is an effective agent for CMRP
of methyl acrylate, which can be initiated by the thermal
decomposition of either a diazo complex or the air-stable
cobalt alkyl complex 2, to afford pMA by a common RT
mechanism.
10 R. Bryaskova, C. Detrembleur, A. Debuigne and R. Je
Macromolecules, 2006, 39, 8263–8268.
´
rome,
11 S. Maria, H. Keneyoshi, K. Matyjaszewski and R. Poli,
Chem.–Eur. J., 2007, 13, 2480–2492.
12 A. Debuigne, C. Michaux, C. Jerome, R. Jerome, R. Poli and
´ ´
C. Detrembleur, Chem.–Eur. J., 2008, 14, 7623–7637.
13 B. B. Wayland, G. Poszmik, S. L. Mukerjee and M. Fyrd, J. Am.
Chem. Soc., 1994, 116, 7943–7944.
The square-planar CoII complex 1 is an effective agent for
the CMRP of methyl acrylate,22 although polydispersities at
longer reaction times are large, suggesting a competitive
catalytic chain transfer process. Polymerization may occur
by either a reversible termination or a degenerative transfer
mechanism, depending on the concentration of available
radicals. The polymerisation may be initiated by the thermal
decomposition of either a diazo complex or 2, the CoIII
derivative of 1, and in either case the reaction initially proceeds
at the same rate, suggesting a common reversible termination
mechanism. Ethyl complex 2 is therefore a rare example of a
well-characterized air-stable five-coordinate cobalt alkyl
complex capable of both initiating and controlling the radical
polymerization process. Studies are underway to probe the
potential effects of temperature, added donor ligands, and
variation in the alkyl and b-ketoaminate substituents on
CMRP efficiency.
14 (a) P. J. McCarthy, R. J. Hovey, K. Ueno and A. E. Martell,
J. Am. Chem. Soc., 1955, 77, 5820–5824; (b) M. J. Carter,
D. P. Rillema and F. Basolo, J. Am. Chem. Soc., 1974, 96,
392–400.
15 (a) S. Z. Haider, A. Hashem, K. M. A. Malik and
M. B. Hursthouse, J. Bangladesh Acad. Sci., 1980, 4, 139–150;
(b) W. T. Robinson and G. A. Rodley, Acta Crystallogr., Sect. A,
1972, 28, S52.
16 C. Ollivier and P. Renaud, Chem. Rev., 2001, 101, 3415–3434.
17 (a) L. G. Marzilli, M. F. Summers, N. Bresciani-Pahor,
E. Zangrando, J.-P. Charland and L. Randaccio, J. Am. Chem.
Soc., 1985, 107, 6880–6888; (b) S. Bruckner, M. Calligaris,
¨
G. Nardin and L. Randaccio, Inorg. Chim. Acta, 1969, 3,
308–312.
1
1
2
18 V-70 t (50 1C) calculated as 37 min, given t (30 1C) = 10 h and
2
6
1
t (60 1C) = 10.2 min .
2
19 In this and other experiments under RT conditions, longer (>8 h)
reaction times exhibit rates that begin to slow dramatically after
approximately 40% MA conversion, in part due to reduced MA
concentration. Compound 1 is not sufficiently soluble in MA for
controlled polymerization in the absence of solvent.
We thank both NSERC of Canada and UBC Okanagan for
funding.
20 A. A. Gridnev and S. D. Ittel, Chem. Rev., 2001, 101,
3611–3659.
21 This result stands in contrast to polymerisations initiated by
porphyrin cobalt alkyl complexes13 or polymer chains capped with
Co(acac)2,8,12 which exhibit linear kinetic behaviour at their outset.
This illustrates the important role of relative strengths of the Co–C
bonds in the dormant polymer chain vs. the initial alkyl initiator.
22 That square planar 1 is an effective CMRP agent for methyl
acrylate is consistent with previous results that show square planar
Co porphyrin compounds control polymerisation of acrylates but
effectively inhibit that of vinyl acetate (VOAc) under RT condi-
tions.3,5 Similarly, treatment of 1 with V-70 and VOAc at 60 1C in
toluene ([1]i : [V-70]i : [VOAc]i = 1 : 0.6 : 500) affords only 18%
conversion and Mn = 3000 after 8 d.
Notes and references
z Compound 1: 1H NMR (CDCl3 25 1C) d 24.7 (2H), 14.2 (2H), 7.7
(4H), 1.7 (4H), ꢃ20.7 (6H), ꢃ78.5 (2H); magnetic moment (CDCl3):
1.99 mB; anal. calcd for C22H22N2O2Co (found): C 65.19 (64.88),
H 5.47 (5.58), N 6.91 (6.65)%.
y Compound 2: 1H NMR (CDCl3 25 1C) d 7.93 (m 4H), 7.35 (m 6H),
5.80 (s 2H), 3.56 (m 4H), 3.39 (q 2H), 2.13 (s 12H), ꢃ0.16 (t 3H); anal.
calcd for C24H27N2O2Co (found): C 66.36 (66.04), H 6.26 (6.33),
N 6.26 (6.08)%.
ꢀc
This journal is The Royal Society of Chemistry 2010
2458 | Chem. Commun., 2010, 46, 2456–2458