Synthesis of end-functional poly(vinyl acetate) by cobalt-mediated radical polymerization

Article abstract of DOI:10.1021/ma047726q

PolyCvinyl acetate) (PVAc) chains prepared by cobalt-mediated radical polymerization in the presence of cobalt(II) acetylacetonate (Co(acac) ₂) were quenched by radical scavengers, such as thiol compounds and nitroxides, to displace the covalently bonded Co(acac)₂ moiety and to end-cap them with a reactive group. The cobalt complex was completely removed by filtration, as confirmed by the induction coupled plasma (ICP) analysis of the polymer before and after treatment. Growing poly(vinyl acetate) chains can be end-functionalized either by addition of an appropriately functionalized nonpolymerizable olefin or by displacement of the Co(acac)₂ moiety by a functionalized nitroxide. This strategy allows PVAc to be synthesized with a predictable molecular weight, a reasonably low polydispersity (M ₂/M_n ～ 1.1-1.3), and a functional ω end group, e.g., hydroxyl and epoxy.

Full text of DOI:10.1021/ma047726q

5452
Macromolecules 2005, 38, 5452-5458
Synthesis of End-Functional Poly(vinyl acetate) by Cobalt-Mediated
Radical Polymerization
Antoine Debuigne,† Jean-Raphae1l Caille,‡ and Robert Je´roˆme*,†
Center for Education and Research on Macromolecules (CERM), University of Lie`ge, Sart-Tilman, B6,
4000 Lie`ge, Belgium, and Solvay Research and Technology, rue de Ransbeek 310,
B-1120 Brussels, Belgium
Received November 5, 2004; Revised Manuscript Received February 24, 2005
ABSTRACT: Poly(vinyl acetate) (PVAc) chains prepared by cobalt-mediated radical polymerization in
the presence of cobalt(II) acetylacetonate (Co(acac)2) were quenched by radical scavengers, such as thiol
compounds and nitroxides, to displace the covalently bonded Co(acac)2 moiety and to end-cap them with
a reactive group. The cobalt complex was completely removed by filtration, as confirmed by the induction
coupled plasma (ICP) analysis of the polymer before and after treatment. Growing poly(vinyl acetate)
chains can be end-functionalized either by addition of an appropriately functionalized nonpolymerizable
olefin or by displacement of the Co(acac)2 moiety by a functionalized nitroxide. This strategy allows PVAc
to be synthesized with a predictable molecular weight, a reasonably low polydispersity (Mw/Mn ∼ 1.1-
1.3), and a functional ω end group, e.g., hydroxyl and epoxy.
Scheme 1
Introduction
Nowadays, a variety of polymers with well-defined
molar masses and low polydispersity can be synthesized
by the controlled radical polymerization (CRP) pro-
cesses, mainly by nitroxide-mediated polymerization
(NMP),1 atom transfer radical polymerization (ATRP),2,3
and radical addition-fragmentation transfer (RAFT).4
Similarly to living ionic polymerizations, telechelic
polymers can be prepared by controlled radical polym-
erization which has the advantage of being tolerant of
many nonprotected functional groups (amine, alcohol,
epoxy). The R-end group is controlled by the structure
of the initiator, whereas the control agent has a decisive
effect on the ω-end group and its possible derivatization
into a desired organic function. Examples of chain-end
functionalization have been reported by RAFT,5 by
NMP,6 and more extensively by ATRP.3,7,8
Scheme 2
impose the structure of the ω-end groups or, merely, to
remove the cobalt complex originally attached to the
chains. In this work, three scavengers were tested in
the cobalt-mediated radical polymerization of vinyl
acetate, i.e., alkanethiol, nitroxide, and nonpolymeriz-
able olefins bearing different reactive groups.
This paper deals with the chain-end functionalization
of poly(vinyl acetate) (PVAc) prepared by a recently
reported controlled radical polymerization based on a
cobalt complex.9 In this CRP method, cobalt acetyl-
acetonate reacts reversibly with the poly(vinyl acetate)
chains, which are accordingly involved in an equilibrium
between active and dormant species (Scheme 1). After
polymerization, all of the chains are end-capped by a
cobalt(II) acetylacetonate (Co(acac)2) moiety. Such a
mechanism has already been proposed for the control
of acrylate polymerization in the presence of cobalt
porphyrin10 and cobaloxime complexes.11
The energy of the cobalt-carbon bond is known to be
low enough for homolysis to occur at moderate temper-
ature.12 Kinetics of homolysis of the Co-C bond has
been analyzed by using radical traps, such as thiols13
and nitroxides,14 which can scavenge the radicals as
soon as they are released by the homolytic cleavage of
the cobalt-carbon bond (Scheme 2). The appropriate
choice of the scavenger is thus a direct way either to
Experimental Section
Materials. Vinyl acetate (VAc) (>99%, Acros) was dried
over calcium hydride, degassed by several freeze-thawing
cycles before being distilled under reduced pressure, and stored
under argon. Methanol was degassed by bubbling argon for
30 min. Toluene and THF were distilled from sodium/ben-
zophenone, followed by bubbling with argon for 30 min. Argon
was also bubbled through 1-propanethiol (99%, Aldrich),
3-butene-1-ol (96%, Aldrich), 1,2-epoxy-5-hexene (97%, Al-
drich), and 3-butenoic acid (97%, Aldrich) for 20 min. Cobalt-
(II) acetylacetonate (Co(acac)2, >98%, Merck), 2,2′-azobis(4-
methoxy-2,4-dimethylvaleronitrile) (V-70, t1/2 ) 10 h at 30 °C)
(Wakko), 2,2,6,6-tetramethylpiperidine 1-oxy (TEMPO) (98%,
Aldrich), 4-hydroxy-2,2,6,6-tetramethylpiperidine 1-oxyl (>97%,
Fluka), sodium hydride (NaH, 60%, dispersion in mineral oil),
and epichlorohydrin (99%, Aldrich) were used as received.
Characterization. Size exclusion chromatography (SEC)
was carried out in THF (flow rate: 1 mL min-1) at 40 °C using
a Waters 600 liquid chromatrograph equipped with a 410
refractive index detector and styragel HR columns (four
columns HP PL gel 5 µm 105, 104, 103, 102 Å). A calibration
curve obtained from polystyrene standards was used to
determine the molecular weights. 1H NMR spectra were
recorded with a Bruker AM 400 spectrometer (400 MHz) in
deuterated chloroform. Induction coupled plasma (ICP) atomic
emission spectroscopy was carried out with a spectrometer
† University of Lie`ge.
‡ Solvay Research and Technology.
* To whom correspondence should be addressed: Tel (32)4-
3663565; Fax (32)4-3363497; e-mail rjerome@ulg.ac.be.
10.1021/ma047726q CCC: $30.25 © 2005 American Chemical Society
Published on Web 05/25/2005

Macromolecules, Vol. 38, No. 13, 2005
Synthesis of End-Functional Poly(vinyl acetate) 5453
(3510 ICP Bausch & Lomb). Samples were prepared by
dissolving a small amount (few milligrams) of poly(vinyl
acetate) in 1 mL of HNO3 (65%) at 60 °C for 2 h. These
solutions were diluted with 9 mL of bidistilled water at room
temperature. An external calibration was necessary for de-
termining the cobalt content. Infrared spectra (IR) were
recorded with a Perkin-Elmer FT-IR instrument (KBr). El-
ementary analyses (EA) were carried out with a Carlo-Erba
elemental analyzer (CHNS-O EA1108).
Polymerization of Vinyl Acetate. V-70 (200 mg, 6.5 ×
10-4 mol) and Co(acac)2 (52 mg, 2 × 10-4 mol) were added into
a 30 mL flask and degassed by three vacuum-argon cycles.
Vinyl acetate (10.0 mL, 108 × 10-3 mol) was then added with
a syringe under argon. The purple mixture was stirred and
heated at 30 °C. After a few hours, the solution changed from
purple to dark green. No polymerization occurred for several
hours, followed by a substantial increase in the solution
viscosity. After 23.5 h at 30 °C, the monomer conversion was
determined by weighing the polymer collected upon removal
of the unreacted monomer in vacuo at 50 °C (conversion )
20%). The mixture was then diluted in acetone, (re)precipitated
in a ice/water mixture, filtered, and dried in vacuo. The green
poly(vinyl acetate) (1.6 g) was analyzed by SEC (Mn,theor ) 9300
g/mol, Mn,SEC ) 8000 g/mol, Mw/Mn ) 1.20).
Treatment of a PVAc End-Capped by Cobalt Complex
with 1-Propanethiol. Green poly(vinyl acetate) (1.6 g, Mn,SEC
) 8000 g/mol, Mw/Mn ) 1.20), prepared according to the
general procedure, was added to a 50 mL flask under an inert
atmosphere and diluted with 10 mL of degassed methanol.
Then, degassed 1-propanethiol (4 × 10-3 mol, 0.4 mL) was
added with a syringe under an inert atmosphere, and the
medium was stirred at 50 °C. After a few hours, the medium
changed from green to dark black, and stirring was maintained
for 24 h. After elution through celite (removal of a black), the
polymer was (re)precipitated in heptane before being dried in
vacuo at 40 °C. A slightly yellow poly(vinyl acetate) (1.0 g)
was recovered and analyzed by SEC and ICP (Mn,SEC ) 8000
g/mol, Mw/Mn ) 1.20).
of VAc) after 47 h at 30 °C. Before injection of the unsaturated
compound, a sample was withdrawn (conversion ) 40%, Mn,SEC
) 13 000 g/mol, Mw/Mn ) 1.17). After 2 h, the monomer
conversion did not change anymore (42%) as result of the
polymerization inhibition. After dilution by acetone, PVAc was
(re)precipitated two times in heptane. A green polymer (0.28
g) was collected by filtration and dried in vacuo at 60 °C.
Mn,theor ) 11 800 g/mol, Mn,SEC ) 14 000 g/mol, Mw/Mn ) 1.20,
Mn,NMR,R ) 86.09 × (3 × (-CH-OCOCH3, PVAc)/(-OCH3,
V-70)) ) 10 500 g/mol, Mn,NMR,ω ) 86.09 × (3 × (-CH-
OCOCH3, PVAc)/(-CH- and -CH2 epoxy, ω-end group)) )
9800 g/mol (functionalization slightly higher than 100%, cf.
1
supra). H NMR: 2.85 ppm (-CH- epoxy, ω-chain end); 2.71
and 2.44 ppm (-CH2 epoxy, ω-chain end).
3-Butenoic Acid. Degassed 3-butenoic acid (4 × 10-3 mol)
was also added to the vinyl acetate polymerization medium
(3.25 × 10-4 mol of V-70, 1.0 × 10-4 mol of Co(acac)2, 54 ×
10-3 mol of VAc) in order to end-functionalize the PVAc.
TEMPO. A degassed solution of TEMPO (2.0 × 10-4 mol)
in toluene (1.0 mL) was added to the polymerization medium
(6.5 × 10-4 mol of V-70, 1.0 × 10-4 mol of Co(acac)2, 54 × 10-3
mol of VAc, 16 h, 26% VAc conversion) and reacted for 2 h at
30 °C. After dilution by acetone, the solution was eluted
through acidic alumina oxide in order to eliminate a purple
deposit of Co(acac)2. PVAc was (re)precipitated two times in
heptane and dried at 60 °C in vacuo to yield 0.47 g of a colorless
polymer. Mn,theor ) 12 100 g/mol, Mn,SEC ) 16 500 g/mol, Mw/
Mn ) 1.15, Mn,NMR,R ) 86.09 × (3 × (-CH-OCOCH3, PVAc)/
(-OCH3, V-70)) ) 12 000 g/mol, Mn,NMR,ω ) 86.09 × ((-CH-
OCOCH3, PVAc)/(AcO-CH-ON-, ω-end group)) ) 18 300
g/mol. (65% functionalization). 1H NMR: 6.20 ppm (AcO-CH-
ON-), 1.6-1.0 ppm (partially hidden -CH2- and CH3 of
TEMPO at ω-chain end).
4-(Glycidyloxy)-2,2,6,6-tetramethylpiperidine 1-oxyl was syn-
thesized as follows. NaH (0.75 g, 60% in mineral oil, 19 × 10-3
mol) was added to a solution of 4-hydroxy-2,2,6,6-tetrameth-
ylpiperidine 1-oxyl, (2.6 g, 15 × 10-3 mol), in THF (15 mL).
The mixture was stirred at room temperature for 30 min under
argon and evaporated under reduced pressure. Epichlorohy-
drin (10.0 g, 108 × 10-3 mol) was added to the residue, and
the mixture was stirred at 50 °C for 20 h under argon. Excess
of epichlorohydrin was distilled off under reduced pressure.
The residual red oil was purified by column chromatography
(chloroform/silica gel). The collected product (1.3 g) was then
purified by distillation under reduced pressure. A vermilion
solid was recovered (1.2 g, yield ) 35%); mp ) 38.3-38.4 °C.
IR (KBr) νmax: 3481.63; 3044.65; 2976.37; 2928.58; 2867.13;
1605.36; 1462.48; 1374.03 (‚ON); 1360.42; 1241.35 (epoxy);
1217.53; 1176.71; 1091.66; 911.35; 894.34; 850.12; 799.09;
761.66 cm-1. Elementary analysis: Calcd for C12H22O3N: C,
63.13%; H, 6.40%; N, 9.71%. Found: C, 63.09%; H, 6.31%; N,
10.03%.
A degassed solution of 4-(glycidyloxy)-2,2,6,6-tetramethylpi-
peridine 1-oxyl in toluene (1.0 mL, 3.0 × 10-4 mol/mL) was
added to the polymerization medium (6.5 × 10-4 mol of V-70,
1.0 × 10-4 mol of Co(acac)2, 54 × 10-3 mol of VAc, 15.5 h, 12%
VAc conversion) and reacted for 1 h at 30 °C. After dilution
by acetone, the solution was eluted through acidic alumina
oxide. The polymer was (re)precipitated two times in heptane
and dried at 60 °C in vacuo to yield 0.25 g of colorless PVAc.
Mn,theor ) 5600 g/mol, Mn,SEC ) 5700 g/mol, Mw/Mn ) 1.16,
Mn,NMR,R ) 86.09 × (3 × (-CH-OCOCH3, PVAc)/(-OCH3,
V-70)) ) 5800 g/mol, Mn,NMR,ω ) 86.09 × ((-CH-OCOCH3,
PVAc)/(AcO-CH-ON-, ω-end group)) ) 7200 g/mol (80%
functionalization). 1H NMR: 6.18 ppm (AcO-CH-ON-); 3.6-
3.3 ppm (-CH-O-CH2-, at ω-chain end); 3.09 ppm (-CH-,
epoxy); 2.77 and 2.58 ppm (-CH2, epoxy), 1.6-1.0 (partially
hidden -CH2- and CH3 of the nitroxyl group at ω-chain end).
Termination of the Vinyl Acetate Polymerization by
Radical Scavengers. 1-Propanethiol. In an example, de-
gassed 1-propanethiol (2 × 10-3 mol, 0.2 mL) was added at 30
°C to the polymerization medium (3.25 × 10-4 mol of V-70,
1.0 × 10-4 mol of Co(acac)2, 54 × 10-3 mol of VAc, 22 h, 14%
VAc conversion), which was diluted with methanol 1 h later
and eluted through celite in order to remove a black deposit.
The polymer was precipitated two times in heptane and dried
at 60 °C in vacuo. A very slightly brown polymer was recovered
(0.17 g); Mn,theor ) 6500 g/mol, Mn,SEC ) 8900 g/mol, Mw/Mn )
1.15, Mn based on the R-end group (Mn,NMR,R ) 86.09 × (3 ×
(-CH-OCOCH3, PVAc)/(-OCH3, V-70)) ) 7200 g/mol, Mn
based on the ω-end group (Mn,NMR,ω) ) 86.09 × (2 × (-CH-
OCOCH3, PVAc)/(-CH2-OCOCH3, ω-end group)) ) 8000 g/mol
1
(90% functionalization). H NMR (CDCl3): 4.84 ppm (-CH-
OCOCH3, backbone); 4.04 ppm (-CH2-OCOCH3, tail); 3.14
ppm (-OCH3, V-70); 2.00 ppm (-OCO-CH3, backbone); 1.85-
1.6 ppm (-CH2-CH-OCOCH3, backbone); 1.5-1.2 ppm (-CH3
of V-70).
3-Butene-1-ol. Degassed 3-butene-1-ol (0.34 g, 0.4 mL, 4.6
× 10-3 mol) was added to the polymerization medium (3.25 ×
10-4 mol of V-70, 1.0 × 10-4 mol of Co(acac)2, 54 × 10-3 mol of
VAc) after 21 h at 30 °C. The monomer conversion (17%) did
not change anymore as a result of the polymerization inhibi-
tion. After dilution by acetone, PVAc was (re)precipitated two
times in ice/water bath. A green polymer (0.20 g) was recovered
by filtration and dried in vacuo. Mn,theor ) 8400 g/mol; Mn,SEC
) 10 000 g/mol; Mw/Mn ) 1.30; Mn,NMR,R ) 86.09 × (3 × (-CH-
OCOCH3, PVAc)/(-OCH3, V-70))
)
10 500 g/mol,
Mn,NMR,ω ) 86.09 × (2 × (-CH-OCOCH3, PVAc)/(-CH2-OH,
ω-end group)) ) 8300 g/mol. In this case, functionalization
would exceed 100%, which is the consequence of the low
intensity of the NMR resonances and the related unaccuracy.
1H NMR (CDCl3): 3.62 ppm (-CH2-OH, ω-chain end).
1,2-Epoxy-5-hexene. Degassed 1,2-epoxy-5-hexene (4 × 10-3
mol, 0.5 mL) was added to the polymerization medium (3.25
× 10-4 mol of V-70, 1.0 × 10-4 mol of Co(acac)2, 32 × 10-3 mol
Results and Discussion
A. 1-Propanethiol as Scavenger. It was previously
reported that the radical polymerization is under control
(linear dependence of molar mass on monomer conver-
sion, low polydispersity) when initiated by 2,2′-azobis-

5454 Debuigne et al.
Macromolecules, Vol. 38, No. 13, 2005
Table 1. Induction Coupled Plasma (ICP) Measurement of the Cobalt Content of PVAc before and after Treatment with
1-Propanethiol
PVAca
aspect
PVAc in solution
(mg/mL)
cobalt in solution
cobalt content in
PVAc (ppm)
(µg/mL)
Mn,theor (g/mol)
blank
before treatment
after treatment
0
2.32
2.55
0.0033
15.58
e0.0033
green
colorless
6715.5
e1.3
8800b
a Mn,SEC ) 8000 g/mol. b Theoretical molar mass (Mn,theor) of PVAc-Co has been calculated from the weight ratio of PVAc over Co
(Mn,theor (g/mol) ) (MPVAc (g/mL) × 58.93 (g/mol))/MCo (g/mL)).
Scheme 3
Scheme 4
(4-methoxy-2,4-dimethylvaleronitrile) (V-70) in the pres-
ence of Co(acac)2 at 30 °C (Scheme 1).9 However, the
final polymer was collected with a green color, typical
of alkyl-cobalt(III) complex, even after precipitation in
heptane and water and after elution through alumina
oxide and silica gel. The coloration persistency against
all these posttreatments is evidence that the cobalt
complex is chemically bonded to the polymer chains.
Nevertheless, it is essential to eliminate this complex
because of the toxicity of cobalt complexes.
In the early 1970s, the chemistry of alkyl-cobalt
complexes was actively investigated.15 The thermal
homolytic cleavage of the Co-C bond in the presence
of an alkanethiol (R′SH) in methanol resulted in the
protonation of the released carbon centered radical, as
shown in Scheme 3.13,16 This potential method of releas-
ing the cobalt derivative from the polymer was tested
on a green PVAc end-capped by the Co(acac)2 complex
(Mn,SEC ) 8000 g/mol, Mw/Mn ) 1.2), prepared by
polymerization of vinyl acetate initiated by V-70 at 30
°C in the presence of Co(acac)2. 1-Propanethiol was
added to a green methanol solution of PVAc-Co, fol-
lowed by filtration through celite and precipitation of
polymer from the filtrate in heptane. An almost colorless
polymer was recovered, consistent with cleavage of the
covalent cobalt-carbon bond. Moreover, SEC chromato-
grams of PVAc before and after treatment with 1-pro-
panethiol (Figure 1) show that the PVAc radicals,
released by homolysis at 30 °C, reacted preferentially
with 1-propanethiol rather than by recombination.
Indeed, no chain coupling was observed after the post-
treatment with the thiol.
in Table 1. The efficiency of an alkanethiol to eliminate
the Co complex of the PVAc chains is highlighted by an
ICP signal, which is at the lower limit of detection and
leads to the conclusion that there is less than 1.3 ppm
of residual Co. Moreover, the experimental molar mass
of the untreated poly(vinyl acetate) (Mn,SEC ) 8000
g/mol) is quite consistent with the molar mass calculated
from cobalt content before treatment.
The same reaction was then realized in situ by adding
and excess of degassed 1-propanethiol during the course
of the vinyl acetate bulk polymerization in the presence
of Co(acac)2 (monomer conversion ) 14%), the trapping
of the growing poly(vinyl acetate) radicals being the
expected result (Scheme 4). After stirring for 1 h at 30
°C, the polymerization medium turned from green to
dark black. A black insoluble matter was eliminated by
filtration, whereas the polymer (re)precipited from the
filtrate is quasi-colorless (Mn,SEC ) 8900 g/mol, Mw/Mn
) 1.15). Loss of green coloration is evidence that PVAc
free of cobalt complex, which was confirmed by 1H NMR
analysis that shows a proton at the PVAc ω-chain end
rather than Co(acac)2 (Figure 2). In addition to major
resonances of the main-chain repeat units ((a) 4.8 ppm,
The Co content of PVAc was measured by the induc-
tion coupled plasma (ICP) atomic emission spectroscopy
before and after treatment with alkanethiol as reported
Figure 2. 1H NMR spectrum of poly(vinyl acetate) (Mn ) 7200
g/mol) after addition of 1-propanethiol (3.25 × 10-4 mol of V-70,
1.0 × 10-4 mol of Co(acac)2, and 54 × 10-3 mol of VAc at 30
°C, followed by addition of 2 × 10-3 mol of propanethiol).
Figure 1. Size exclusion chromatograms of poly(vinyl acetate)
(0.2 × 10-3 mol, Mn,SEC ) 8000 g/mol) before (dotted line) and
after addition of 1-propanethiol (4 × 10-3 mol) (full line).

Macromolecules, Vol. 38, No. 13, 2005
Synthesis of End-Functional Poly(vinyl acetate) 5455
Scheme 5
Figure 3. Plot of ln([M]0/[M]) vs time for the vinyl acetate
(54 × 10-3 mol) polymerization initiated by V-70 (3.25 × 10-4
mol) at 30 °C in the presence of cobalt(II) acetylacetonate (1.0
× 10-4 mol), with (2) and without (b) addition of 3-butene-1-
ol (4.6 × 10-3 mol).
1H, -CH-OCOCH3; (b) 2.0 ppm, 3H, -OCO-CH3; (c)
1.85-1.60 ppm, 2H, -CH2-CH-OCOCH3), signals
characteristic of the initiator fragment are observed.
The methoxy protons -OCH3 (d) and the methyl
protons (e, f) of V-70 are noted respectively at 3.14 and
1.5-1.2 ppm, whereas the signal of the methylene
protons (g) is more likely hidden by the signal for the
methylene protons of the chain (c). The molar mass of
the chains can be determined from the intensity ratio
of the resonance (d) (three protons) for the initiator
fragment and resonance (a) (one proton) for the main
chain [Mn,NMR(R-end) ) 86.09 × (3 × (a/d)) ) 7200 g/mol].
The low intensity signal (h), at 4.04 ppm (h/d ) 1.8/3),
has been assigned to methylene protons of the ω-end
group, consistent with the substitution of a proton for
the cobalt complex (Mn,NMR(ω-end) ) 86.09 × (2 × (a/h))
) 8000 g/mol, 90% functionalization).
Addition of a radical scavenger, such as an al-
kanethiol, to growing PVAc chains is not only an
efficient technique for eliminating metal contaminants,
but it also paves the way to the end-functionalization
of the chains as discussed hereafter.
Figure 4. 1H NMR spectrum of poly(vinyl acetate) added with
3-butene-1-ol (3.25 × 10-4 mol of V-70, 1.0 × 10-4 mol of
Co(acac)2, and 54 × 10-3 mol of VAc at 30 °C, followed by
addition of 4.6 × 10-3 mol of 3-butene-1-ol).
B. Nonpolymerizable Olefins as Scavengers. One
technique of chain end-capping in controlled radical
polymerization consists of adding nonpolymerizable
olefins, such as 1,2-epoxy-5-hexene and allyl alcohol, to
the polymerization medium. In the case of ATRP of
acrylates, the growing chains were deactivated by these
nonpolymerizable olefins. Epoxy and hydroxyl functions
were accordingly attached as ω-end groups together
with a halogen atom.8 This strategy has been extended
to the cobalt-mediated radical polymerization of VAc.
Three nonpolymerizable olefins, i.e., 3-butene-1-ol, 1,2-
epoxy-5-hexene, and 3-butenoic acid, have been tested
in order to end-functionalize PVAc according to Scheme
5. In the case of success, the chains will be irreversibly
capped by Co(acac)2 and could not be reactivated for
subsequent monomer addition according to the cobalt-
mediated vinylation, largely reported by Giese,17 Pat-
tenden,18 and Branchaud.19
protons (a, b, c), signals for the initiator (V-70) fragment
at R-chain end (d, e, f, g) and a signal (h) for the CH2-
OH protons at ω-chain end (3.62 ppm) were observed
(Figure 4) [Mn,NMR(R-end) ) 10 500 g/mol, Mn,NMR(ω-end)
) 8300 g/mol]. Within the limits of the NMR accuracy
(very low intensity of the NMR peaks characteristic of
the ω-end group), the functionalization is close to
completion. No signal corresponding to an unsaturated
chain end was detected, which confirms the absence of
dehydrocobaltation reaction.
Addition of 1,2-Epoxy-5-hexene. Similarly, 1,2-epoxy-
5-hexene has been added to the growing chains. Once
again, the polymer chains rapidly lost their capacity to
grow further, the monomer conversion leveling off at
42% in this specific example. After purification by (re)-
precipitation in heptane, a polymer with Mn,SEC
)
14 000 g/mol and Mw/Mn ) 1.20 was collected and
analyzed by 1H NMR spectroscopy. Protons of PVAc (a,
b, c) and protons of the initiator V-70 at the R-chain
end (d, e, f, g) was observed together with CH (h) and
CH2 (i, j) protons of the epoxy group at 2.85, 2.71, and
2.44 ppm, respectively (Figure 5). The end-capping
efficiency has been confirmed by the good agreement
between molar masses determined considering the
methylene protons at the ω-chain end (9800 g/mol) and
the methoxy protons at the R-chain end (10 500 g/mol).
Functionalization is also close to 100% (cf. supra).
Addition of 3-Butene-1-ol. An excess of 3-butene-1-ol
was added to growing poly(vinyl acetate) chains, initi-
ated by V-70 at 30 °C in the presence of Co(acac)2, at
low monomer conversion. The vinyl acetate polymeri-
zation was accordingly inhibited, the monomer conver-
sion remaining constant at 17% in this example. Figure
3 shows a sharp deviation of ln([M]0/[M]) vs time from
the ideal linear dependence evolution. A polymer with
low molar mass and low polydispersity was recovered
by repeated precipitation in a ice/water bath (Mn,SEC
)
1
10 000 g/mol, Mw/Mn ) 1.30). H NMR confirmed that
3-butene-1-ol was incorporated as an end group. Indeed,
in addition to the signals characteristic of the PVAc
Addition of 3-Butenoic Acid. 3-Butenoic acid was used
as a nonpolymerizable substituted olefin in order to end-

5456 Debuigne et al.
Macromolecules, Vol. 38, No. 13, 2005
Scheme 6
lent with respect to Co(acac)2) to the VAc polymerization
medium, at low monomer conversion (26%), results in
the chain growth termination and formation of a purple
deposit characteristic of the Co(acac)2 complex. This
insoluble compound was eliminated by elution through
alumina column. PVAc was twice precipitated in hep-
tane, and a colorless polymer has been collected (Mn,SEC
) 16 500 g/mol, Mw/Mn ) 1.15). The displacement of the
cobalt complex at the chain ends by TEMPO reaction
has been confirmed not only by the loss of the green
Figure 5. 1H NMR spectrum of poly(vinyl acetate) added with
1,2-epoxy-5-hexene (3.25 × 10-4 mol of V-70, 1.0 × 10-4 mol
of Co(acac)2, and 32 × 10-3 mol of VAc at 30 °C, followed by
addition of 4 × 10-3 mol of 1,2-epoxy-5-hexene).
1
color but also by H NMR and ICP analysis.
Table 2. Addition of 3-Butenoic Acid to the
Cobalt-Mediated Radical Polymerization of VAc
The 1H NMR signals characteristic of the CH2 (j, from
1.5 to 1.2 ppm) and CH3 (i, from 1.2 to 1.0 ppm) protons
of TEMPO are observed although in partial overlap with
the signals for the protons of the initiator V-70 (e, f, g
from 1.6 to 1.2 ppm) (Figure 6). Moreover, the proton
at the junction between PVAc and TEMPO (Ac-O-
CH-ON-, h, 6.20 ppm) is clearly detected. However,
the intensity ratio of the methoxy protons d of V-70 and
the proton h does fit not exactly prediction based on the
chain end-capping by TEMPO (65% functionalization
within the limits of the experimental error). Indeed, the
intensity of the proton h is too low to estimate the molar
mass of the polymer with precision.
time (h) conv (%) Mn,SEC (g/mol) Mn,theor (g/mol) Mw/Mn
19a
5
15
3800
8400
2300
7000
1.20
1.72
21b
a Addition of the acid. b 2 h after the addition of 3-butenoic acid.
cap PVAc with a carboxylic acid group. However,
addition of this olefin to the polymerization medium
does not prevent the chains from growing further, as
assessed by the vinyl acetate conversion that increased
from 5% to 15%, 2 h after 3-butenoic acid was added.
Nevertheless, the control on the vinyl acetate polymer-
ization was perturbed, as shown by the broadening of
the molecular weight distribution from 1.20 to 1.72
(Table 2). From the 1H NMR analysis of the end groups
of the chains, there is no evidence that the olefin has
been added as ω-end group, in contrast to the initiator
fragment that can be detected. A too low reactivity
might be the reason for lack of addition, whereas
interference of the carboxylic acid with the cobalt
complex could explain a loss in the polymerization
control.
The completeness of the cobalt complex extraction
from the poly(vinyl acetate) has been estimated by the
ICP analysis of Co before and after addition of TEMPO
to PVAc. Data are listed in Table 3. There is a good
agreement between the apparent molar mass of the
In conclusion, the addition of nonpolymerizable ole-
fins, such as 3-butene-1-ol and 1,2-epoxy-5-hexene, to
cobalt-mediated polymerization of vinyl acetate leads
to well-defined end-functional poly(vinyl acetate). If not
properly chosen, the reactive substituent of the olefin
may prevent it from adding and possibly interfere with
the polymerization mechanism, as observed in the case
of carboxylic acid. Whenever the end-capping is effec-
tive, the cobalt catalyst remains attached to the chains,
although it could be eliminated by subsequent treat-
ment with 1-propanethiol, if necessary.
C. Nitroxide Scavengers. Addition of TEMPO.
Nitroxides, including TEMPO, are potential trapping
agent of macroradicals. The cobalt-mediated radical
polymerization of vinyl acetate has been tentatively
quenched by TEMPO with the purpose to displace the
cobalt complex from the chain end (Scheme 6). Addition
of a degassed toluene solution of TEMPO (two equiva-
Figure 6. 1H NMR spectrum of poly(vinyl acetate) added with
2,2,6,6-tetramethylpiperidine 1-oxy (TEMPO) (6.5 × 10-4 mol
of V-70, 1.0 × 10-4 mol of Co(acac)2, and 54 × 10-3 mol of VAc
at 30 °C, followed by addition of 2.0 × 10-4 mol of TEMPO).

Macromolecules, Vol. 38, No. 13, 2005
Synthesis of End-Functional Poly(vinyl acetate) 5457
Table 3. Induction Coupled Plasma (ICP) Measurement of the Cobalt Content of PVAc before and after Addition of
TEMPO
PVAca
aspect
PVAc in solution
(mg/mL)
cobalt in solution
cobalt content in
PVAc (ppm)
(µg/mL)
Mn,theor (g/mol)
blank
PVAc-Co
PVAc-TEMPO
0
3.1
3.0
0.0033
9.995
0.0033
green
colorless
3255.7
<1.1
18300b
a Mn,SEC ) 16 500 g/mol. b Theoretical molar mass (Mn,theor) of PVAc-Co has been calculated from the weight ratio of PVAc over Co
(Mn,theor (g/mol) ) (MPVAc (g/mL) × 58.93 (g/mol))/MCo (g/mL)).
Scheme 7
untreated poly(vinyl acetate) measured by SEC (16 500
g/mol) and the molar mass calculated on the basis of
the cobalt content measured by ICP (18 300 g/mol). The
efficacy of the substitution by TEMPO has to be found
in a residual content of cobalt as low as 1.1 ppm after
treatment by TEMPO.
The thermal stability of the covalent bonding of
TEMPO to PVAc explains why the radical polymeriza-
tion of VAc cannot be controlled by nitroxide-mediated
polymerization.20 Advantage of this characteristic fea-
ture can be taken for the end-functionalization of cobalt
free PVAc.
Figure 7. 1H NMR spectrum of poly(vinyl acetate) added with
4-(glycidyloxy)-2,2,6,6-tetramethylpiperidine 1-oxyl (6.5 × 10-4
mol of V-70, 1.0 × 10-4 mol of Co(acac)2, and 54 × 10-3 mol of
VAc at 30 °C, followed by addition of 3.0 × 10-4 mol of
4-(glycidyloxy)-2,2,6,6-tetramethylpoperidine 1-oxyl).
Addition of 4-(Glycidyloxy)-2,2,6,6-tetramethylpiperi-
dine 1-oxyl. A functional group bearing nitroxide should
be an effective end-capping agent. To confirm this
expectation, 4-(glycidyloxy)-2,2,6,6-tetramethylpiperi-
dine 1-oxyl was synthesized according to a procedure
reported by Endo et al.,21 i.e., by reaction of 4-hydroxy-
2,2,6,6-tetramethylpiperidine 1-oxyl with epichlorohy-
drin (Scheme 7). This epoxy-containing nitroxyl radical
was added to the vinyl acetate polymerization medium
(actual monomer conversion of 10%), resulting in the
termination chains (unchanged VAc conversion). The
solution collected after elution through an alumina
column was colorless, and PVAc precipitated in heptane
was analyzed by SEC (Mn,SEC ) 5700 g/mol, Mw/Mn )
1.16) and 1H NMR (Figure 7). In addition to the signals
for the protons of the PVAc main chain (a, b, c) and for
fragment of V-70 (d, e, f, g), signal corresponding to the
piperidyl group (i, j) and signals characteristic of the
epoxy group (k, l, m) are observed as well as the -CH-
O-CH2- protons (n, o) and the proton (h) at the
junction between PVAc and the nitroxyl group (AcO-
CH-ON-) (Figure 7). End-functionalization efficiency
has been calculated by NMR from the ratio of the molar
masses determined on the basis of the protons h and
d, respectively (80% efficiency). In this case, a partial
opening of the epoxy end groups might contribute to
decrease the functionalization efficiency. Therefore,
termination of the VAc polymerization by a functional
nitroxide turns out to be an effective strategy for the
end-capping of cobalt free chains.
as a solid compound which is then easily removed by
filtration. Nonpolymerizable olefins substituted by an
alcohol or an epoxy group are also terminating agents
which are incorporated to the chains as an end group.
The cobalt complex remains however covalently bonded
to the PVAc chains. More interestingly, nitroxides, e.g.
TEMPO, also stop the growth of the chains as result of
irreversible substitution of the Co(acac)2. On this basis,
the addition of a functional group containing nitroxide
to the growing PVAc chains is an efficient strategy to
prepare well-defined, end-functional, and cobalt free
PVAc by a one-pot technique.
Acknowledgment. The authors gratefully acknowl-
edge Solvay for financial support and fellowship to A.D.
They are grateful to A. Momtaz and V. Bodart for
fruitful discussions. They also thank Wakko for the gift
of V-70. They are indebted to the “Belgian Science
Policy” for financial support in the frame of the “Inter-
university Attraction Poles Programme (PAI V/03)-
Supramolecular Chemistry and Supramolecular Cataly-
sis”.
References and Notes
(1) (a) Moad, G. H.; Solomon, D. H.; Johns, S. R.; Willing, R. I.
Macromolecules 1982, 15, 1188. (b) Georges, M. K.; Veregin,
R. P. N.; Kazmaier, P. M.; Hamer, G. K. Macromolecules
1993, 26, 2987. (c) Hawker, C. J.; Bosman, A. W.; Hart, E.
Chem. Rev. 2001, 101, 3661.
(2) (a) Kato, M.; Kamigaito, M.; Sawamoto, M.; Higashimura,
T. Macromolecules 1995, 28, 1721. (b) Wang, J. S.; Matyjas-
zewski, K. J. Am. Chem. Soc. 1995, 117, 5614. (c) Kamigaito,
M.; Ando, T.; Sawamoto, M. Chem. Rev. 2001, 101, 3689.
Conclusions
This paper has reported on methods for the end-
functionalization of poly(vinyl acetate) prepared by
cobalt-mediated radical polymerization. Addition of an
alkanethiol to the polymerization medium terminates
the chains by a proton and releases the cobalt complex

5458 Debuigne et al.
Macromolecules, Vol. 38, No. 13, 2005
(3) Matyjaszewski, K.; Xia, J. Chem. Rev. 2001, 101, 2921.
(4) (a) Chiefari, J.; Chong, Y. K.; Ercole, F.; Kristina, J.; Jeffery,
J.; Le, T. P. T.; Mayadunne, R. T. A.; Meijs, G. F.; Moad, C.
L.; Moad, G.; Rizzardo, E.; Thang, S. H. Macromolecules 1998,
31, 5559. (b) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.;
Chong, Y. K.; Moad, G.; Thang, S. H. Macromolecules 1999,
32, 6977. (c) Moad, G.; Chiefari, J.; Chong, Y. K.; Krstina, J.;
Mayadunne, R. T. A.; Postma, A.; Rizzardo, E.; Thang, S. H.
Polym. Int. 2000, 49, 993.
(5) Lai, J. T.; Filla, D.; Shea, R. Macromolecules 2002, 35, 6754.
(6) (a) Rodlert, M.; Harth, E.; Rees, I.; Hawker, C. J. J. Polym.
Sci., Polym. Chem. 2000, 38, 4749. (b) Hawker, C. J.; Hedrick,
J. L. Macromolecules 1995, 28, 2993. (c) Harth, E.; Hawker,
C. J.; Fan, W.; Waymouth, R. M. Macromolecules 2001, 34,
3856.
(7) (a) Coessens, V.; Pintauer, T.; Matyjaszewski, K. Prog. Polym.
Sci. 2001, 26, 337. (b) Haddleton, D. M.; Waterson, C.
Macromolecules 1999, 32, 8732. (c) Coessens, V.; Pyun, J.;
Miller, P. J.; Gaynor, S. G.; Matyjaszewski, K. Macromol.
Rapid Commun. 2000, 21, 103.
(8) Coessens, V.; Pyun, J.; Miller, P. J.; Gaynor, S. G.; Matyjas-
zewski, K. Macromol. Rapid Commun. 2000, 21, 103.
(9) Debuigne, A.; Caille, J.-R.; Je´roˆme, R. Angew. Chem., Int. Ed.
2005, 44, 1101-1104.
(12) Halpern, J. Science 1985, 227, 869.
(13) Tsou, T. T.; Loots, M.; Halpern, J. J. Am. Chem. Soc. 1982,
104, 623.
(14) Flora, T. T. Ng.; Remple, G. L.; Mancuso, C.; Halpern, J.
Organometallics 1990, 9, 2762.
(15) (a) Kemmitt, R. D. W.; Russel, D. R. In Comprehensive
Organometallic Chemistry; Wilkinson, G., Stone, A., Abel, E.,
Eds.; Pergamon Press: Oxford, 1982; Vol. 5, pp 1-276. (b)
Sweany, R. L. In Comprehensive Organometallic Chemistry
II; Wilkinson, G., Stone, A., Abel, E., Eds.; Pergamon Press:
Oxford, 1995; Vol. 8, pp 1-114.
(16) McAllister, R.; Weber, J. H. J. Organomet. Chem. 1974, 77,
91.
(17) Ghosez, A.; Go¨bel, T.; Giese, B. Chem. Ber. 1988, 121, 1807.
(18) (a) Branchaud, B. P.; Meier, M. S.; Choi, Y. Tetrahedron Lett.
1988, 29, 167. (b) Branchaud, B. P.; Choi, Y. L. Tetrahedron
Lett. 1988, 29, 6037. (c) Branchaud, B. P.; Yu, G.-X. Orga-
nometallics 1993, 12, 4262.
(19) (a) Patel, V. F.; Pattenden, G. J. Chem. Soc., Chem. Commun.
1987, 871. (b) Bandaranayake, W. M.; Pattenden, G. J. Chem.
Soc., Chem. Commun. 1988, 1179. (c) Howell, A. R.; Pat-
tenden, G. J. Chem. Soc., Perkin Trans. 1 1990, 2715. (d)
Coveney, D. J.; Patel, V. F.; Pattenden, G. Tetrahedron Lett.
1987, 28, 5949.
(20) Lutz, J.-F.; Lacroix-Demazes, P.; Boutevin, B.; Le Mercier,
C.; Gigmes, D.; Bertin, D.; Tordo, P. Polym. Prepr. (Am.
Chem. Soc., Div. Polym. Chem.) 2002, 43 (2), 287.
(21) Endo, T.; Takuma, K.; Takata, T.; Hirose, C. Macromolecules
1993, 26, 3227.
(10) (a) Wayland, B. B.; Poszmik, G.; Mukerjee, S. L.; Fryd, M. J.
Am. Chem. Soc. 1994, 116, 7943. (b) Wayland, B. B.; Basickes,
L.; Mukerjee, S. L.; Wei, M.; Fryd, M. Macromolecules 1997,
30, 8109. (c) Lu, Z.; Fryd, M.; Wayland, B. B. Macromolecules
2004, 37, 2686.
(11) Arvanitopoulos, L. D.; Greuel, M. P.; King, B. M.; Shim, A.
K.; Harwood: H. J. ACS Symp. Ser. 1998, 685, 316-331.
MA047726Q