Searching for more effective agents and conditions for the RAFT polymerization of MMA: Influence of dithioester substituents, solvent, and temperature

Article abstract of DOI:10.1021/ma0480650

A series of tertiary ditbiobenzoates differently substituted on the pbenyl group (Z) were synthesized in order to investigate the RAFT-mediated polymerization of MMA. The chain-transfer agent 2-cyanoprop-2-yl dithiobenzoate, although an excellent RAFT agent for polymerization of MMA, does not yield a very narrow polydispened polymer at low conversion. The introduction of electron-withdrawing substituents on the dithiobenzoate aromatic ring improves the activity of the chain-transfer agent narrowing the molecular weight distribution, especially in the first 2 h of the process; conversely, electron-donating substituents deteriorate the performance of the process in the early stages. Substituents on the dithiobenzoate that, for reasons of steric hindrance reduce the conjugation of the phenyl with the C-S double bond of the dithiogroup, diminish the effectiveness of the chain-transfer agent. The nature of the leaving group (R) also has a strong influence on the process: the substitution of 2-cyanoprop-2-yl with the bulkier 2-cyano-4-methylpent-2-yl as leaving group improves remarkably the MMA polymerization, especially in the early stages. The solvent effect on MMA RAFT polymerization using benzene, acetonitrile, and DMF was also investigated. The lower propagation rate constant of MMA in benzene gives rise to reduced rates of polymerization but narrower polydispersities of the polymer, especially at low conversions. The RAFT polymerizations of MMA were carried out at 60 and 90°C to assess the temperature dependence of the process. The higher temperature increases both the rate of polymerization and the transfer constant of the RAFT agent, resulting in lower polydispersities at a given conversion. Ab initio calculations confirmed the observed experimental results.

Full text of DOI:10.1021/ma0480650

Macromolecules 2005, 38, 3129-3140
3129
Searching for More Effective Agents and Conditions for the RAFT
Polymerization of MMA: Influence of Dithioester Substituents, Solvent,
and Temperature
Massimo Benaglia,#,* Ezio Rizzardo,‡ Angelo Alberti,# and Maurizio Guerra#
Istituto per la Sintesi Organica e la Fotoreattivita` (ISOF), Consiglio Nazionale delle Ricerche (CNR),
via Piero Gobetti, 101 - 40129 Bologna, Italy, and Commonwealth Scientific & Industrial Research
Organisation (CSIRO), Molecular Science, Clayton VIC 3168, Australia
Received September 20, 2004; Revised Manuscript Received January 10, 2005
ABSTRACT: A series of tertiary dithiobenzoates differently substituted on the phenyl group (Z) were
synthesized in order to investigate the RAFT-mediated polymerization of MMA. The chain-transfer agent
2-cyanoprop-2-yl dithiobenzoate, although an excellent RAFT agent for polymerization of MMA, does
not yield a very narrow polydispersed polymer at low conversion. The introduction of electron-withdrawing
substituents on the dithiobenzoate aromatic ring improves the activity of the chain-transfer agent
narrowing the molecular weight distribution, especially in the first 2 h of the process; conversely, electron-
donating substituents deteriorate the performance of the process in the early stages. Substituents on the
dithiobenzoate that, for reasons of steric hindrance reduce the conjugation of the phenyl with the CdS
double bond of the dithiogroup, diminish the effectiveness of the chain-transfer agent. The nature of the
leaving group (R) also has a strong influence on the process: the substitution of 2-cyanoprop-2-yl with
the bulkier 2-cyano-4-methylpent-2-yl as leaving group improves remarkably the MMA polymerization,
especially in the early stages. The solvent effect on MMA RAFT polymerization using benzene, acetonitrile,
and DMF was also investigated. The lower propagation rate constant of MMA in benzene gives rise to
reduced rates of polymerization but narrower polydispersities of the polymer, especially at low conversions.
The RAFT polymerizations of MMA were carried out at 60 and 90 °C to assess the temperature dependence
of the process. The higher temperature increases both the rate of polymerization and the transfer constant
of the RAFT agent, resulting in lower polydispersities at a given conversion. Ab initio calculations
confirmed the observed experimental results.
Introduction
initiate the radical process. However, the polydispersity
of the polymer obtained at low conversion is higher than
that observed using CTAs characterized by higher
transfer constant (e.g., bearing 2-cyanoprop-2-yl as
leaving group). The group Z in dithioesters 2 influences
the stability of the intermediate radicals and modifies
the reactivity of CdS double bond toward radical
addition.5,8,15 The group Z plays an important role
especially in MMA polymerization and, for instance, a
phenyl as a stabilizing Z group is preferable in order to
narrow the polydispersity and to prevent a bimodal
molecular weight distribution.16 The faster consumption
of the initial RAFT agent 2 is due to the role of the
phenyl group which increases the stability of the
intermediate radicals 3 and 4 and thereby increases the
addition rate of the growing radical chains to the CdS
double bond.5 However, even 2-cyanoprop-2-yl dithioben-
zoate (2a), considered one of the best CTAs for MMA
polymerization, does not give a very narrow polydis-
persed polymer at low conversion.
In the present work, new RAFT agents were synthe-
sized varying the substituents on the phenyl of the
dithiobenzoate in order to probe steric, polar, and
resonance effects on the RAFT process. To achieve this,
methyl-, methoxy-, fluoro-, cyano-, trifluoromethyl-, and
phenyl-substituted dithiobenzoates were investigated as
MMA polymerization mediators. Aromatic Z groups
such as pyridine and naphthalene were also considered.
The effects of different substituent groups on the
molecular weight distribution of the polymer, especially
in the first 2 h of the polymerization were observed. The
influence of the substitution position on the aromatic Z
The reversible addition fragmentation chain-transfer
process (RAFT) has been intensively investigated since
its first reports.1-3 The reason for the great interest in
this methodology resides in its applicability to a wide
variety of monomers, ability to afford polymers of low
polydispersity, ability to finely control the molecular
weight,4,5 and ability to allow for the generation of
complex macromolecular architectures such as gradient
or block copolymers2,3 and star polymers.4,6,7 By choosing
the appropriate RAFT agent, it is possible to fit the
different monomer requirements.2,8-11 Among the more
common monomers, MMA, whose polymeric growing
chain is a tertiary radical, is one of the most demanding
with regard to the chain-transfer agent (CTA). Thus,
in the dithioester 2 (Scheme 1),12 R must be an excellent
leaving group but, at the same time, a good reinitiating
radical. Indeed, the tertiary cumyl and 2-cyanoprop-2-
yl are appropriate for MMA polymerization because they
suit both of the essential requirements, whereas the
triphenylmethyl radical, being too stable, cannot initiate
a new radical chain and the benzyl group does not have
an adequate leaving ability.5,13 Recently a polymeriza-
tion of MMA controlled by a CTA which has a secondary
leaving R group (methoxycarbonylphenylmethyl) was
reported.14 This radical appears to be sufficiently stable
for replacement by the PMMA growing chain and, at
the same time, reactive enough toward MMA to re-
* Author to whom correspondence should be addressed. E-
mail: m.benaglia@isof.cnr.it.
‡ CSIRO - Mol. Sci. Victoria.
# ISOF - CNR Bologna.
10.1021/ma0480650 CCC: $30.25 © 2005 American Chemical Society
Published on Web 03/22/2005

3130 Benaglia et al.
Macromolecules, Vol. 38, No. 8, 2005
Scheme 1. Steps Involving the Dithioester Functionality in the RAFT Polymerization Mechanism
7.62 (m, p-ArH, 2H); 8.09 (d, J ) 8.6 Hz, o-ArH, 4H). 13C NMR
group was also studied, as well as that of the bulkiness
(100 MHz, CDCl3), δ: 127.58 (ArC-3); 128.66 (ArC-2); 133.18
of the leaving group. The solvent and temperature
(ArC-4); 143.66 (ArC-1); 219.54 (C)S).
dependence of the process were also investigated. The
Bis(2,6-dimethylthiobenzoyl) Disulfide (1b). Following
most representative intermediate radical models were
Method A using 2-bromo-m-xylene (1.33 mL) as starting
subjected to ab initio calculations, and the calculated
material, the crude product was purified by flash chromatog-
enthalpy barrier of the addition and the fragmentation
raphy (eluent, hexane/methylene chloride ) 8:2). A crystalline
reactions of the intermediate radicals were compared
red solid (yield 91.8%) was obtained. 1H NMR (200 MHz,
with the experimental results.
CDCl3), δ: 2.33 (s, CH3, 12H); 7.05-7.28 (m, ArH, 6H). 13C
NMR (50 MHz, CDCl3), δ: 19.53 (CH3); 128.01 (ArC-3); 129.17
(ArC-2); 133.87 (ArC-4); 144.01 (ArC-1); 229.05 (CdS).
Experimental Section
Chemicals. 2-Bromo-m-xylene (Aldrich, 98%), 4-bromo-m-
xylene (Aldrich, 97%), 4-bromoanisole (Aldrich, 99%), 1-bromo-
2,4-dimethoxybenzene (Aldrich, 97%), 1-bromo-4-fluorobenzene
(Aldrich, 99%), 1-bromo-2,4-difluorobenzene (Aldrich, 98%),
R-bromo-p-tolunitrile (Aldrich, 98%), 5-methylisophthalonitrile
(Lancaster, 97%), 3,5-bis(trifluoromethyl)benzyl bromide (Al-
drich, 97%), 2,3,4,5,6-pentafluorobenzyl chloride (Lancaster,
97%), 4-bromobiphenyl (Aldrich, 98%), 2-bromonaphthalene
(Aldrich, 97%), 1-(bromomethyl)naphthalene (Aldrich, 98%),
4-(chloromethyl)pyridine hydrochloride (Fluka, 97%), 1 N
iodine solution (Aldrich, 0.96N in water), sodium methoxide
solution (Aldrich, 25% in methanol), carbon disulfide (Fluka,
99%), and 2,2′-azobis(2,4-dimethylpentanenitrile) (VAZO52,
Du Pont, 99%) were used as received. Azobis(isobutirronitrile)
(AIBN, Aldrich 98%) was crystallized from methanol. p-
Toluenesulfonic acid monohydrate (Aldrich, 98.5%) was an-
hydrified in a Dean-Stark apparatus using toluene as azeo-
tropic solvent. Methyl methacrylate (MMA, Aldrich, 99%) was
purified by filtration through alumina, fractionated under
reduced pressure, and flash distilled immediately prior to use.
THF for synthesis was freshly distilled from LiAlH4. Silica gel
(Kieselgel 60, Merck, 230-400 mesh) was used for flash
chromatography purification.
Bis(2,4-dimethylthiobenzoyl) Disulfide (1c). Following
Method A using 4-bromo-m-xylene (1.35 mL) as starting
material, the crude product was purified by flash chromatog-
raphy (eluent, hexane/methylene chloride ) 8:2). A red liquid
(yield 75.3%) was obtained. 1H NMR (200 MHz, CDCl3), δ: 2.38
(s, CH3, 6H), 2.40 (s, CH3, 6H); 7.04-7.38 (m, ArH, 6H). 13C
NMR (50 MHz, CDCl3), δ: 19.72 (o-CH3); 21.30 (p-CH3);
[126.25; 126.89; 131.85; 134.46; 140.50; 142.13 (ArC)]; 222.78
(CdS).
Bis(4-methoxythiobenzoyl) Disulfide (1d). Following
Method A using 4-bromoanisole (1.25 mL) as starting material,
the crude product was purified by chromatography on alumina
(eluent, methylene chloride). A deep red solid (yield 87%) was
obtained. 1H NMR (400 MHz, CDCl3), δ: 3.90 (s, CH3, 6H);
6.93 (d, J ) 8.8 Hz, m-ArH, 4H); 8.19 (d, J ) 8.8 Hz, o-ArH,
4H). 13C NMR (100 MHz, CDCl3), δ: 55.67(CH3); 113.84 (ArC-
3); 130.06 (ArC-2); 136.91 (ArC-1); 164.23 (ArC-4); 217.56(Cd
S).
Bis(2,4-dimethoxythiobenzoyl) Disulfide (1e). Follow-
ing Method A using 1-bromo-2,4-dimethoxybenzene (2.17 g)
as starting material, the crude product (1.34 g) was not
purified because of its instability toward cleaning procedures
and was directly used for the subsequent reaction.
Analytical Techniques. NMR analysis were performed on
a Bru¨ker AC200 (1H ) 200 MHz, 13C ) 50 MHz), Bru¨ker
DRX500 (1H ) 500 MHz, 13C ) 125.6 MHz), or Varian Mercury
400 (1H ) 400 MHz, 13C ) 100 MHz) spectrometers in CDCl3
(δ, ppm from TMS). Mass Spectroscopy data were collected
on a ThermoQuest MAT95XL high-resolution mass spectrom-
eter. Size exclusion chromatography (SEC) was performed on
a Waters Associates liquid chromatograph equipped with a
differential refractometer and with three PLgel, 5 µm, mixed
C columns (MW range 200-2 000 000) and one PLgel, 3 µm,
mixed E column (MW range up to 30 000) at 22 °C with THF
as eluent (flow rate 1 mL/min) calibrated with Polymer
Laboratories polystyrene standards.
Bis(thiobenzoyl) Disulfide (1a). (Method A). A THF
solution (10 mL) of bromobenzene (1.05 mL, 10 mmol) was
added dropwise to a mixture of Mg (0.25 g, 10 mg atom) in
THF (40 mL) and a crystal of I2 in a two-necked round-bottom
flask equipped with condenser and a magnetic stirrer with a
nitrogen atmosphere. To initiate the reaction, the mixture was
gently heated during the initial stages of the addition. After 3
h, carbon disulfide (1 mL, 16.7 mmol) was added dropwise and
the solution was stirred for further 3 h. The solvent was then
removed under vacuum and the resulting deep red viscous
liquid was dissolved in a diluted K2CO3 solution (100 mL),
filtered, washed with ethyl ether (2 × 50 mL), and poured in
a round-bottom flask equipped with magnetic stirrer. An
aqueous solution of iodine 0.96 N (10.42 mL, 10 mmol) was
then added dropwise. During the addition, the solution started
to change color from dark red to pink as the disulfide
precipitated. After elimination of excess I2 with few crystals
of Na2S2O3, the mixture was extracted with methylene chlo-
ride, dried over sodium sulfate, and evaporated. The crude
product was used for the subsequent reaction without any
further purification. A crystalline red solid (yield 90.2%) was
obtained. 1H NMR (400 MHz, CDCl3), δ: 7.46 (m, m-ArH, 4H);
Bis(4-fuorothiobenzoyl) Disulfide (1f). Following Method
A using 1-bromo-4-fluorobenzene (1.1 mL) as starting material,
the crude product was purified by flash chromatography
(eluent, hexane/methylene chloride ) 8:2). A deep red solid
(yield 45%) was obtained. 1H NMR (200 MHz, CDCl3), δ: 7.07-
7.23 (m, m-ArH, 4H); 8.08-8.22 (m, o-ArH, 4H). 13C NMR (50
2
MHz, CDCl3) δ: 115.77 (d, JCF ) 22 Hz, 3-ArC); 130.00 (d,
1
3JCF ) 9.3 Hz 2-ArC);139.84 (ArC-1); 165.96 (d, JCF ) 255.5
Hz, 4-ArC); 217.20 (CdS).
Bis(2,4-difluorothiobenzoyl) disulfide (1g). Following
Method A using 1-bromo-2,4-difluorobenzene (1.13 mL) as
starting material, the crude product (0.66 g) was not purified
because of its instability toward cleaning procedures and was
directly used for the subsequent reaction.
Bis(4-cyanothiobenzoyl) Disulfide (1i). (Method B). A
25% solution of sodium methoxide in methanol (4.57 mL, 20
mmol) was added to a suspension of sulfur (0.64 g, 20 mmol)
and R-bromo-p-tolunitrile (1.96 g, 10 mmol) in methanol (50
mL) in a round-bottom flask equipped with a condenser and
magnetic stirring. After 3 h, the solvent was removed and the
resulting dark brown crude product was dissolved in a diluted
solution of NaHCO3. After being washed with ethyl ether (2
× 50 mL), an aqueous solution of iodine 0.96 N (10.42 mL, 10
mmol) was added dropwise. During the addition, the solution
started to change color from dark brown to pink as the
disulfide precipitated. After elimination of excess I2 with few
crystals of Na2S2O3, the mixture was extracted with ethyl ether
and dried over sodium sulfate. After removal of the solvent,
the brown solid crude product (yield 85.4%) was used for the
subsequent reaction without any further purification. 1H NMR
(400 MHz, CDCl3), δ: 7.76 (d, J ) 8 Hz, o-ArH, 4H); 8.11 (d,
J ) 8 Hz, m-ArH, 4H). 13C NMR (100 MHz, CDCl3) δ: 116.17
(ArC-4); 117.80 (CN); 127.89 (ArC-2); 132.50 (ArC-3); 146.11
(ArC-1); 216.23 (CdS).

Macromolecules, Vol. 38, No. 8, 2005
RAFT Polymerization of MMA 3131
Bis(3,5-dicyanothiobenzoyl) Disulfide (1j). Following
Method B using 3,5-dicyano R-bromotoluene17 (2.07 g) as
starting material, the crude product (0.44 g) was not purified
because of its instability toward cleaning procedures and was
directly used for the subsequent reaction.
6H, C(CH3)2CN); 2.33 (s, 3H, o-ArCH3); 2.36(s, 3H, p-ArCH3);
6.98-7.21 (m, 3H, ArH). 13C NMR (50 MHz, CDCl3), δ: 19.40
(o-ArCH3); 21.24 (p-ArCH3); 26.09 (C(CH3)2CN); 41.97 (C(CH3)2-
CN); 119.81 (CN); [126.31; 127.71; 131.853; 133.75; 140.08;
143.19 (ArC)]; 228.05 (CdS). Accurate Mass: 249.0650 (theo-
retical mass for C13H15NS2: 249.0646).
2-Cyanoprop-2-yl(4-methoxy) Dithiobenzoate (2d). Fol-
lowing the procedure for compound 2b using 1d (0.513 g) and
AIBN as azo compound, after flash chromatography (eluent,
methylene chloride), a crystalline red solid (yield 86%) was
obtained. 1H NMR (200 MHz, CDCl3), δ: 1.94 (s, 6H, C(CH3)2-
CN); 3.87 (s, 3H, OCH3); 6.88 (d, J ) 8.95 Hz, 2H, m-ArH);
8.00 (d, J ) 8.95, 2H, o-ArH). 13C NMR (50 MHz, CDCl3), δ:
26.66 (C(CH3)2CN); 41.50 (C(CH3)2CN); 55.61 (OCH3); 120.29
(CN); 113.66 (ArC-3); 128.77 (ArC-2); 137.66 (ArC-1); 164.02
(ArC-4); 220.40 (CdS). Accurate Mass: 251.0441 (theoretical
mass for C12H12NOS2: 251.0439).
2-Cyanoprop-2-yl(2,4-dimethoxy) Dithiobenzoate (2e).
Following the procedure for compound 2b using 1e (0.597 g)
and AIBN as azo compound, after flash chromatography
(eluent, ethyl ether), a red solid (yield 79.8%) was obtained.
1H NMR (200 MHz, CDCl3), δ: 1.87 (s, 6H, C(CH3)2CN); [3.80
(s, 3H); 3.82 (s, 3H) (OCH3)]; 6.35-6.53 (m, 2H, m-ArH); 7.69
(d, J ) 8.6, 1H, o-ArH). 13C NMR (50 MHz, CDCl3), δ: 26.16
(C(CH3)2CN); 41.48 (C(CH3)2CN); [55.40; 55.64 (OCH3)]; 120.37
(CN); [98.27; 105.14; 128.03; 131.93; 156.92; 163.53 (ArC)];
221.05 (CdS). Accurate Mass: 281.0541 (theoretical mass for
C13H15NO2S2: 281.0544).
Bis(3,5-bis(trifluoromethyl)thiobenzoyl) Disulfide (1k).
Following Method B using 3,5-bis(trifluoromethyl)benzyl bro-
mide (1.83 mL) as starting material, the crude product (0.87
g) was not purified because of its instability toward cleaning
procedures and was directly used for the subsequent reaction.
Bis(2,3,4,5,6-pentafluorothiobenzoyl) Disulfide (1l).
Following Method B using 2,3,4,5,6-pentafluorobenzyl chloride
(1.36 mL) as starting material, the crude product was not
purified (yield 22.2%) and was directly used for the subsequent
reaction. 13C NMR (100 MHz, CDCl3) δ: 118.23 (tt, 2JCF ) 16.6
Hz, 3JCF ) 3.9 Hz, ArC-1); 137.72 (dm, 1JCF ) 257.0 Hz, ArC-
1
1
3); 142.34 (dm, JCF ) 257.9 Hz, ArC-2); 146.58 (dm, JCF
244.0 Hz, ArC-4); 209.67 (CdS).
)
Bis(4-phenylthiobenzoyl) Disulfide (1n). Following
Method A using 4-bromobiphenyl (2.33 g) as starting material,
after the addition of the iodine solution, the aqueous suspen-
sion was filtered and washed with water and then with
methanol. A light brown solid (yield 92%) poorly soluble in
cold solvents (ethyl acetate, chloroform) was obtained. 1H NMR
(400 MHz, CDCl3/CS2: 1/1), δ: [7.38-7.49 (m, 6H); 7.63-7.68
(m, 8H) (ArH)]; 8.20 (d, J ) 8 Hz, 4H, p-ArH). 13C NMR (100
MHz, CDCl3/CS2: 1/1) δ: [126.92; 126.97; 128.20; 128.86;
139.36; 142.03; 145.65 (ArC)]; 217.29 (CdS).
Bis(2-naphthylthioyl) Disulfide (1o). Following Method
A using 2-bromo-naphthalene (2.07 g) as starting material, the
crude product (0.86 g) was not purified because of its instability
toward cleaning procedures and was directly used for the
subsequent reaction.
Bis(1-naphthylthioyl) Disulfide (1p). Following Method
B using 1-(bromomethyl)naphthalene (2.21 g) as starting
material, the crude product (1.34 g) was not purified because
of its instability toward cleaning procedures and was directly
used for the subsequent reaction.
2-Cyanoprop-2-yl(4-fluoro) Dithiobenzoate (2f). Fol-
lowing the procedure for compound 2b using 1f (0.479 g) and
AIBN as azo compound, after flash chromatography (eluent,
hexane/ethyl ether ) 7:3), a deep pink crystalline solid (yield
92%) was obtained. 1H NMR (200 MHz, CDCl3), δ: 1.94 (s,
6H, C(CH3)2CN); 7.08 (t, JHF ) 8.55 Hz, 2H, m-ArH); 7.97 (m,
2H, o-ArH). 13C NMR (50 MHz, CDCl3), δ: 26.47 (C(CH3)2CN);
2
45.85 (C(CH3)2CN); 115.60 (d, JCF ) 22 Hz, ArC-3); 119.85
(CN); 128.99 (d, 3JCF ) 8.85 Hz, ArC-2); 140.84 (ArC-1); 165.85
1
(d, JCF ) 254.2 Hz, ArC-4); 220.80 (CN). Accurate Mass:
Triphenylmethyl Dithioisonicotinate (5h). A solution
of 4-picolyl chloride hydrochloride (1.64 g, 10 mmol) in
methanol (30 mL) was added to a suspension of sulfur (0.65
g, 20 mmol) and sodium methoxide 25% (6.5 mL, 30 mmol) in
methanol (20 mL) in a round-bottom flask equipped with
condenser and magnetic stirring. After 3 h, a solution of
triphenyl bromomethane (3.23 g, 10 mmol) in ethyl ether (30
mL) was added dropwise to the dark-brown mixture that was
stirred for an additional 2 h. The solvent was removed under
vacuum, and the residue was dissolved with ethyl ether and
then filtered. After the solvent was removed, the crude product
was quickly passed through a small column filled with silica
gel using ethyl ether as solvent to remove part of the impuri-
ties. A purple solid (1.7 g) was obtained, and it was used
without further purification because of its instability.
2-Cyanoprop-2-yl Dithiobenzoate (2a). 2a was synthe-
sized as previously reported.12
2-Cyanoprop-2-yl(2,6-dimethyl) Dithiobenzoate (2b).
The solution of 1b (0.508 g, 1.4 mmol) and the azo compound
(AIBN) (0.266 g, 1.62 mmol) in ethyl acetate (120 mL) was
degassed in a round-bottom flask equipped with condenser and
magnetical stirring and then refluxed in nitrogen atmosphere
for 16 h. The end of the reaction was checked by TLC. After
the solvent was removed in a vacuum, the crude product was
purified by flash chromatography (eluent, hexane/ethyl ether
) 7:3) obtaining a crystalline pink solid (yield 96%). 1H NMR
(200 MHz, CDCl3), δ: 1.95 (s, 6H, C(CH3)2CN); 2.34 (s, 6H,
ArCH3); 6.97-7.24 (m, 3H, ArH). 13C NMR (50 MHz, CDCl3),
δ: 18.93 (ArCH3); 26.03 (C(CH3)2CN); 47.54 (C(CH3)2CN);
119.67 (CN); [128.01; 128.97; 133.32; 144.89 (ArC)]; 230.12 (Cd
S). Accurate Mass: 249.0652 (theoretical mass for C13H15-
NS2: 249.0646).
239.0238 (theoretical mass for C11H10FNS2: 239.0239).
2-Cyanoprop-2-yl(2,4-difluoro) Dithiobenzoate (2g).
Following the procedure for compound 2b using 1g (0.530 g)
and AIBN as azo compound, after flash chromatography
(eluent, petroleum spirit/ethyl ether: 8:2), a red oil (yield
1
31.6%) was obtained. H NMR (500 MHz, CDCl3), δ: 1.92 (s,
6H, C(CH3)2CN); 6.80-6.96 (m, 2H, m-ArH); 7.60-7.69 (m, 1H,
o-ArH). 13C NMR (125.6 MHz, CDCl3), δ: 26.23 (C(CH3)2CN);
42.24 (C(CH3)2CN); 104.79 (dd, 2JCF ) 2JCF ) 25.9 Hz, ArC-3);
111.80 (dd, 2JCF ) 21.7 Hz, 4JCF ) 3.7 Hz, ArC-5); 115.10 (dd,
4
2JCF ) 24.0 Hz, JCF ) 7.9 Hz, ArC-1); 119.52 (CN); 131.38
(dd, 3JCF ) 10.3 Hz, 3JCF ) 2.4 Hz, ArC-6); 158.21 (dd, 1JCF
)
257.4 Hz, 3JCF ) 12.8 Hz, ArC-2); 164.46 (dd, 1JCF ) 255.8 Hz,
3
3JCF ) 11.8 Hz, ArC-4); 216.95 (d, JCF ) 3.0 Hz, CdS).
Accurate Mass: 257.0136 (theoretical mass for C11H9F2NS2:
257.0144).
2-Cyanoprop-2-yl Dithioisonicotinate (2h). The solution
of 5h (1.702 g, 4.28 mmol) and AIBN (0.913 g, 5.56 mmol) in
ethyl acetate (120 mL) was degassed in a round-bottom flask
equipped with condenser and magnetic stirring and then
refluxed for 16 h. The end of the reaction was checked by TLC.
After the solvent was removed under vacuum, the crude
product was purified by flash chromatography (eluent, ethyl
ether). 0.69 g of a red oil (yield 72.3%) were obtained. 1H NMR
(200 MHz, CDCl3), δ: 1.92 (s, 6H, C(CH3)2CN); 7.64 (d, J )
4.48 Hz, 2H, o-ArH); 8.70 (d, J ) 4.48 Hz, 2H, m-ArH). 13C
NMR (50 MHz, CDCl3), δ: 26.14 (C(CH3)2CN); 41.91 (C(CH3)2-
CN); 119.12 (CN); 119.69 (ArC-2); 149.50 (ArC-1); 150.34 (ArC-
3); 220.62 (CdS). Accurate Mass: 222.0288 (theoretical mass
for C11H10N2S2: 222.0285).
2-Cyanoprop-2-yl 4-Cyanodithiobenzoate (2i). Follow-
ing the procedure for compound 2b using 1i (0.499 g) and
AIBN as azo compound, the crude product was first purified
by sublimation to remove the AIBN decomposition byproduct
(tetramethyl succinonitrile) and then by flash chromatography
(eluent, hexane/ethyl acetate ) 8:2). A purple crystalline solid
(yield 35%) was obtained. 1H NMR (200 MHz, CDCl3), δ: 1.97
2-Cyanoprop-2-yl(2,4-dimethyl) Dithiobenzoate (2c).
Following the procedure for compound 2b using 1c (0.508 g)
and AIBN as azo compound, after flash chromatography
(eluent, hexane/ethyl ether ) 8:2), a crystalline red solid (yield
1
90.1%) was obtained. H NMR (200 MHz, CDCl3), δ: 1.91 (s,

3132 Benaglia et al.
Macromolecules, Vol. 38, No. 8, 2005
(s, 6H, C(CH3)2CN); 7.72 (d, J ) 8.38 Hz, 2H, o-ArH); 7.97 (d,
J ) 8.38 Hz, 2H, m-ArH). 13C NMR (50 MHz, CDCl3), δ: 26.28
(C(CH3)2CN); 42.18 (C(CH3)2CN); 115.71 (ArC-4); 117.85 (ArCN);
119.28 (C(CH3)2CN); 127.07 (ArC-3); 132.32 (ArC-2); 146.97
(ArC-1); 220.54 (CdS). Accurate Mass: 246.0280 (theoretical
mass for C12H10N2S2: 246.0285).
2-Cyanoprop-2-yl 3,5-Dicyanodithiobenzoate (2j). Fol-
lowing the procedure for compound 2b using 1j (0.569 g) and
AIBN as azo compound, after flash chromatography (eluent,
methylene chloride), a crystalline red solid (yield 25%) was
obtained. 1H NMR (400 MHz, CDCl3), δ: 1.96 (s, 6H, C(CH3)2-
CN); 8.08 (t, J ) 1.47 Hz, 1H, p-ArH); 8.34 (d, J ) 1.47 Hz,
2H, o-ArH). 13C NMR (100 MHz, CDCl3), δ: 26.23 (C(CH3)2-
CN); 42.74 (C(CH3)2CN); 114.67 (ArC-3); 115.67 (ArCN); 118.71
(C(CH3)2CN); 133.49 (ArC-2); 137.56 (ArC-4); 145.31 (ArC-1);
216.43 (CdS). Accurate Mass: 271.0242 (theoretical mass for
C13H9N3S2: 271.0238).
(ArC)]; 226.38 (CdS). Accurate Mass: 271.0492 (theoretical
mass for C18H15NS2: 271.0489).
2-Cyano-4-methylpent-2-yl dithiobenzoate (2q) Follow-
ing the procedure for compound 2b using 1a (0.429 g) and 2,2′-
azobis(2,4-dimethylpentanenitrile) (0.402 g) as azo compound,
after flash chromatography (eluent, hexane/ethyl acetate )
1
8:2), a red oil (yield 95%) was obtained. H NMR (400 MHz,
CDCl3), δ: [1.07 (d, J ) 6.8 Hz, 3H); 1.12 (d, J ) 7.8 Hz, 3H)
(CH(CH3)2]; [1.88 (dd, 1J ) 14.1 Hz, 2J ) 5.6 Hz, 1H); 2.24
(dd, 1J ) 14.1 Hz, 2J ) 6.6 Hz, 1H) (CH2)]; 1.95 (s, 3H, C(CH3)-
CN); 2.08 (m, 1H, CH2CHMe2);7.39 (dd, J ) 7.8 Hz, J ) 7.4
Hz, 2H, m-ArH); 7.53 (t, J ) 7.4 Hz, 1H, p-ArH); 7.89 (d, J )
7.8 Hz, 2H, o-ArH). 13C NMR (100 MHz, CDCl3), δ: [23.56;
24.00 (CH(CH3)2)]; 24.70 (C(CH3)CN); 25.74 (CH2CHMe2);
45.81 (CMeCN); 46.64 (CH2); 119.39 (CMeCN); 126.59 (ArC-
3); 128.45 (ArC-2); 132.73 (ArC-4); 144.74 (ArC-1); 222.98 (Cd
S). Accurate Mass: 263.0807 (theoretical mass for C14H17-
NS2: 263.0802).
2-Cyanoprop-2-yl 3,5-Bis(trifluoromethyl) Dithioben-
zoate (2k). Following the procedure for compound 2b using
1k (0.810 g) and AIBN as azo compound, after flash chroma-
tography (eluent, cyclohexane/ethyl ether ) 8:2), a red-purple
solid (yield 50.4%) was obtained. 1H NMR (200 MHz, CDCl3):
δ: 1.97 (s, 6H, C(CH3)2CN); 8.04 (s, 1H, p-ArH); 8.29 (s, 2H,
o-ArH). 13C NMR (50 MHz, CDCl3), δ: 26.28 (C(CH3)2CN);
2-Cyano-4-methylpent-2-yl-4-cyano Dithiobenzoate (2r).
Following the procedure for compound 2b using 1i (0.499 g)
and 2,2′-azobis(2,4-dimethylpentanenitrile) (0.402 g) as azo
compound, after flash chromatography (eluent, hexane/ethyl
acetate ) 8:2), a purple oil (yield 48%) was obtained. 1H NMR
(400 MHz, CDCl3), δ: [1.08 (d, J ) 6.4 Hz, 3H); 1.12 (d, J )
6.8 Hz, 3H) (CH(CH3)2)]; [1.88 (dd, 1J ) 14.2 Hz, 2J ) 5.8 Hz,
1H); 2.21 (dd, 1J ) 14.2 Hz, 2J ) 6.6 Hz, 1H) (CH2)]; 2.07 (m,
1H, CH2CHMe2); 1.96 (s, 3H, C(CH3)CN); 7.67 (d, J ) 8.4 Hz,
2H, o-ArH); 7.92 (d, J ) 8.4 Hz, 2H, m-ArH). 13C NMR (100
MHz, CDCl3), δ: [23.51; 23.98 (CH(CH3)2)]; 24.57 (C(CH3)CN);
25.77 (CH2CHMe2); 46.37 (CMeCN); 46.61 (CH2); 115.60 (ArC-
4); 117.89 (ArCN); 118.79 (CMeCN); 127.11 (ArC-2); 132.31
(ArC-3); 147.30 (ArC-1); 220.41 (CdS). Accurate Mass: 288.0746
(theoretical mass for C15H16N2S2: 288.0755).
1
42.49 (C(CH3)2CN); 119.07 (CN); 122.71 (q, JCF ) 271.6 Hz,
2
CF3); 125.42 (ArC-2); 126.58 (ArC-4); 132.14 (q, JCF ) 33.8
Hz, ArC-3); 145.28 (ArC-1); 218.64 (CdS). Accurate Mass:
357.0080 (theoretical mass for C13H9F6NS2: 357.0081).
2-Cyanoprop-2-yl 2,3,4,5,6-Pentafluoro Dithiobenzoate
(2l). Following the procedure for compound 2b using 1l (0.681
g) and AIBN as azo compound, after flash chromatography
(eluent, methylene chloride), a red oil (yield 52%) was obtained.
1H NMR (400 MHz, CDCl3): δ: 1.93 (s, 6H, C(CH3)2CN). 13C
NMR (100 MHz, CDCl3), δ: 25.90 (C(CH3)2CN); 42.75 (C(CH3)2-
2-Cyano-4-methylpent-2-yl 3,5-bistrifluoromethyldi-
thiobenzoate (2s). Following the procedure for compound 2b
using 1k (0.810 g) and 2,2′-azobis(2,4-dimethylpentanenitrile)
(0.402 g) as azo compound, after flash chromatography (eluent,
cyclohexane/ethyl ether ) 8:2), a red-purple oil (yield 52%) was
obtained. 1H NMR (400 MHz, CDCl3), δ: [1.09 (d, J ) 6.8 Hz,
3H); 1.14 (d, J ) 6.8 Hz, 3H) (CH(CH3)2)]; [1.90 (dd, 1J ) 14.1
Hz, 2J ) 5.6 Hz, 1H); 2.24 (dd, 1J ) 14.1 Hz, 2J ) 6.2 Hz, 1H)
(CH2)]; 2.08 (m, 1H, CH2CHMe2);1.98 (s, 3H, C(CH3)CN); 8.03
(s, 1H, p-ArH); 8.27 (s, 2H, o-ArH). 13C NMR (100 MHz,
CDCl3), δ: [23.49; 24.00 (CH(CH3)2)]; 24.55 (C(CH3)CN); 25.84
(CH2CHMe2); 46.62 (CH2); 46.65 (CMeCN); 118.63 (CMeCN);
2
CN); 118.60 (CN); 119.52 (t, JCF ) 17.0 Hz, ArC-1); 137.53
1
1
(dm, JCF ) 255.7 Hz, ArC-3); 142.03 (dm, JCF ) 256.6 Hz,
1
ArC-4); 142.29 (dm, JCF ) 249.0 Hz, ArC-2); 206.36 (CdS).
Accurate Mass: 310.9867 (theoretical mass for C11H6F5NS2:
310.9862).
2-Cyanoprop-2-yl-4-pyridiniumdithiocarboxate 4-Tol-
uenesulfonate Salt (2m). 2m was obtained by dissolving 2h
and an equimolar amount of anhydrous 4-methylbenzene-
sulfonic acid in situ when preparing the MMA stock solution.
2-Cyanoprop-2-yl(4-phenyl) Dithiobenzoate (2n). Fol-
lowing the procedure for compound 2b using 1n (0.642 g) and
AIBN as azo compound, after flash chromatography (eluent,
hexane/ethyl ether ) 8:2), a pink crystalline product (yield
76%) was obtained. 1H NMR (200 MHz, CDCl3), δ: 1.97 (s,
6H, C(CH3)2CN); [7.35-7.57 (m, 3H); 7.57-769 (m, 4H); 8.03
(d, J ) 8.22 Hz, 2H) (ArH)]. 13C NMR (50 MHz, CDCl3), δ:
26.55 (C(CH3)2CN); 41.71 (C(CH3)2CN); 120.04 (CN); [127.12;
127.21; 128.34; 128.99; 139.59; 143.19; 145.736 (ArC)]; 221.97
(CdS). Accurate Mass: 297.0648 (theoretical mass for C18H15-
NS2: 297.0646).
2-Cyanoprop-2-yl-2-naphthyl Dithiolate (2o). Following
the procedure for compound 2b using 1o (0.569 g) and AIBN
as azo compound, after flash chromatography (eluent, cyclo-
hexane/toluene/methylene chloride ) 5:4:1) and crystallization
from methanol, a pale red solid (yield 64%) was obtainined.
1H NMR (200 MHz, CDCl3), δ: 1.98 (s, 6H, C(CH3)2CN); [7.12-
7.32 (m, 1H); 7.45-7.68 (m, 2H); 7.77-8.08 (m, 4H) (ArH)].
13C NMR (50 MHz, CDCl3), δ: 26.34 (C(CH3)2CN); 41.64
(C(CH3)2CN); 119.95 (CN); [123.88; 126.39; 127.03; 127.61;
128.24; 128.35; 129.56; 132.06; 135.26; 141.46 (ArC)]; 222.41
(CdS). Accurate Mass: 271.0496 (theoretical mass for C18H15-
NS2: 271.0489).
1
122.73 (q, JCF ) 272 Hz, CF3);125.41 (ArC-4); 126.62 (ArC-
2); 132.11 (q, 2JCF)33.8 Hz, ArC-3); 145.58 (ArC-1); 218.51 (Cd
S). Accurate Mass: 399.0544 (theoretical mass for C16H15F6-
NS2: 399.0550).
2-Cyano-4-methylpent-2-yl-4-methoxyl Dithiobenzoate
(2t). Following the procedure for compound 2b using 1d (0.513
g) and 2,2′-azobis(2,4-dimethylpentanenitrile) (0.402 g) as azo
compound, after flash chromatography (eluent, toluene), a red
oil (yield 22.8%) was obtained. 1H NMR (400 MHz, CDCl3), δ:
[1.07 (d, J ) 6.8 Hz, 3H); 1.11 (d, J ) 6.4 Hz, 3H) (CH(CH3)2)];
1
2
1
[1.87 (dd, J ) 14.2 Hz, J ) 6.8 Hz, 1H); 2.25 (dd, J ) 14.2
2
Hz, J ) 6.2 Hz, 1H) (CH2)]; 2.07 (m, 1H, CH2CHMe2); 1.95
(s, 3H, C(CH3)CN); 3.86 (s, 3H, OCH3); 6.87 (d, J ) 9 Hz, 2H,
m-ArH); 7.98 (d, J ) 9 Hz, 2H, o-ArH). 13C NMR (100 MHz,
CDCl3), δ: [23.57; 24.02 (CH(CH3)2)]; 24.94 (C(CH3)CN); 25.76
(CH2CHMe2); 45.63 (CMeCN); 46.80 (CH2); 55.56 (OCH3);
113.59 (ArC-3); 119.75 (CMeCN); 128.72 (ArC-2); 137.86 (ArC-
1); 163.88 (ArC-4); 220.35 (CdS). Accurate Mass: 293.0913
(theoretical mass for C15H19NOS2: 293.0908).
MMA Polymerizations at 60 °C. A stock solution of MMA
(7.01 M), AIBN (6.3 × 10-3 M), and RAFT agent (1.12 × 10-2
M) in benzene (or a different solvent when specified) was
prepared. Aliquots (2 mL) were transferred to four ampules,
degassed with three freeze-evacuate-thaw cycles, and sealed.
The ampules were heated at 60 ( 1 °C in a thermostated water
bath for 1, 2, 4, and 16 h. The conversions were estimated by
analyzing a portion of the obtained mixtures by 1H NMR (400
MHz) weighting the -C(O)OCH3 integral of MMA (3.75 ppm)
and PMMA (3.59 ppm). The excess monomer and solvent were
2-Cyanoprop-2-yl-1-naphthyl Dithiolate (2p). Following
the procedure for compound 2b using 1p (0.569 g) and AIBN
as azo compound, after flash chromatography (eluent, cyclo-
hexane/toluene/methylene chloride ) 5:4:1), a red, viscous
1
liquid (yield 85%) was obtained. H NMR (200 MHz, CDCl3),
δ: 1.97 (s, 6H, C(CH3)2CN); [7.38-7.65 (m, 4H); 7.80-7.98 (m,
2H); 8.12-8.27 (m, 1H) (ArH)]. 13C NMR (50 MHz, CDCl3), δ:
26.00 (C(CH3)2CN); 42.22 (C(CH3)2CN); 119.67 (CN); [124.32;
124.48; 126.53; 127.26; 128.20; 129.18; 130.56; 133.62; 143.34

Macromolecules, Vol. 38, No. 8, 2005
RAFT Polymerization of MMA 3133
Scheme 2. General Synthetic Strategy for Tertiary
Dithiobenzoates
Conformational Effects of the Group Z. Previous
studies of the RAFT process demonstrated that best
results on MMA polymerization are achieved with an
activating Z group. Barner-Kowollik et al. evidenced the
bimodal behavior of the MMA polymerization using
cumyl phenyldithioacetate (CPDA) as a RAFT agent16
and suggested that, at low monomer conversion, CPDA
acts as a “conventional” chain-transfer agent working
in nonliving mode, whereas at higher monomer conver-
sion, it works in living mode. Similarly, this hybrid
behavior was also observed in MMA polymerization
mediated by S-methoxycarbonylphenylmethyl methyl-
trithiocarbonate (MCPMT), which has a nonactivated
Z group (-SMe), whereas, S-methoxycarbonylphenyl-
methyl dithiobenzoate (MCPDB), which has phenyl as
Z, provided a more uniform polymerization.14 In this
work, we found that methyl groups in ortho positions
diminish the performance of the CTA with respect to
the parent compound 2-cyanoprop-2-yl dithiobenzoate
(2a). Indeed, MMA polymerization mediated by the 2,6-
dimethyldithiobenzoate derivative (2b) showed a trend
from no control in the first 2 h to a low control at the
end of the process (PD1h 1.67, PD2h 1.70, PD4h 1.62,
PD16h 1.46) in comparison to the polymerization of MMA
in absence of CTA (PD1h 1.61, PD2h 1.65, PD4h 2.05,
PD16h 2.32) (Table 2). The SEC molecular weight
distribution reproduced in Figure 1 clearly shows a
bimodal behavior recalling that of MMA polymerizations
mediated by CTAs not carrying a phenyl as group Z.
removed by evaporation at ambient temperature under vacuum,
and the residues were analyzed directly by GPC.
MMA Polymerizations at 90 °C. A stock solution of MMA
(6.55 M), 1,1-azobis(cyclohexanecarbonitrile) (1.18 × 10-3 M)
and RAFT agent (1.13 × 10-2 M) in benzene was prepared.
Aliquots (2 mL) were transferred to four ampules, degassed
with three freeze-evacuate-thaw cycles, and sealed. The
ampules were heated at 90 ( 1 °C in a thermostated oil bath
for 1, 2, 4, and 16 h and then treated as described for
experiments at 60 °C.
Computational Details. DFT calculations were carried out
with the Gaussian 98 system of programs18 employing the
B3LYP functional19,20 and the valence double-ú basis set
supplemented with polarization d functions on heavy atoms21
(B3LYP/6-31G*). The unrestricted wave function was used for
radical species. Energies (E) were corrected for the zero point
vibrational energies (ZPVE) computed at the B3LYP/6-31G*
level from frequency calculations using a scaling factor of
0.9806 to account for anharmonicity.22 Enthalpies (H) in the
gas phase at 333 °K were calculated from eq 1 using the rigid-
rotor and harmonic-oscillator approximation.
H333 ) E + ZPVE + ∆H0f333
(1)
Thus, in the case of 2b, the phenyl does not act as a
stabilizing Z group showing a behavior similar to alkyl
or thioalkyl substituents. The steric hindrance of the
two methyl groups in ortho positions reduces the
conjugation of the aromatic system with the dithiocar-
boxyl group by preventing its free rotation. On the other
hand, the 2,4-dimethyl-substituted dithiobenzoate de-
rivative 2c leads to a living RAFT polymerization of
MMA, as evidenced in the SEC plot shown in Figure 2.
Nevertheless, the polydispersity at low conversion is
inferior to that obtained with the parent compound 2a
(PD1h 1.58, PD2h 1.42, PD4h 1.22, PD16h 1.12) (Table 2).
The plots of molecular weight versus conversion for the
polymerizations mediated by compounds 2a, 2b, and 2c
are linear for 2a, quasi-linear for 2c, and nonlinear for
2b, whose trend highlights the nonliving character of
the MMA polymerization (Figure 3). (In a clean living
polymerization, the plot of Mn versus conversion should
pass through the zero point.)
This behavior suggests that the methyl in the 2-posi-
tion of the dithiobenzoate hampers the rotation of the
phenyl group, reducing its conjugation with the dithio-
carboxyl group and thereby deactivates the CdS double
bond. Similarly, the position of the dithiocarboxyl group
in naphthalene-derivatized CTAs influences the degree
of control of MMA polymerizations. Thus, the different
polydispersity of the polymer obtained using the 1-naph-
thyldithioate derivative 2p (PD1h 1.47, PD2h 1.29) or the
2-naphthyldithioate 2o (PD1h 1.30, PD2h 1.19) (Table 2),
evidences that the isomer substituted in position 1,
which is less free to rotate, has a lower transfer
constant. Recent studies have led to similar results in
the MMA polymerization mediated by 2-cyanoprop-2-
yl 1-dithionaphthalenate27 and 2-cyanoprop-2-yl 1-dithio-
phenanthrenate.28
Results and Discussion
Synthesis of Chain-Transfer Agents. Tertiary
dithio esters were synthesized by the radical reaction
of bis(thiocarbonyl) disulfide 1 with a tertiary azo
compound23,24 (Scheme 2). Thiocarbonyl disulfides hav-
ing an unsubstituted phenyl ring (1a) (Table 1) or a
phenyl ring substituted with methyl (1b, 1c), methoxy
(1d, 1e), fluoro (1f, 1g) and phenyl (1n), or 2-naphtha-
lene (1o) as Z group were synthesized by reacting the
appropriate Grignard reagent with carbon disulfide and
subsequent oxidation of the dithiocarboxylate salt with
aqueous iodine. The aryl Grignard reagents bearing
strongly electron-withdrawing substituents do not react
easily with carbon disulfide to give the corresponding
dithiocarboxylate salt; thus, phenyl derivatives substi-
tuted with cyano (1i, 1j), trifluoromethyl (1k), and the
pentafluorophenyl derivative (1l) were obtained by the
oxidation of the appropriate benzyl bromide or chloride
precursors with sodium methoxide in the presence of
sulfur25 followed by oxidation of the dithiocarboxylate
salt with iodine. Also, 2-cyanoprop-2-yl 1-naphthyldithio-
ate (2p) was obtained in better yields by oxidation of
1-chloromethylnaphthalene in the presence of sulfur
and subsequent reaction of the disulfide product with
AIBN.
It was also difficult to obtain the 4-pyridine derivative
(2-cyanoprop-2-yl dithioisonicotinate) 2h (Table 1) from
the related bis(thiocarboxy) disulfide because the oxida-
tion of the corresponding sodium dithiocarboxylate
failed to occur. Thus, triphenylmethyl dithioisonicoti-
nate was synthesized by reaction of sodium dithioisoni-
cotinate, obtained by the oxidation of 4-picolyl chloride
in the presence of sulfur, with triphenylmethyl chloride
followed by a radical reaction with AIBN (Scheme 3).26
Some of the most significant dithiobenzoates were
synthesized with 2-cyano-4-methylpent-2-yl as leaving
group by reacting disulfides 1 (Scheme 2) with 2,2′-
azobis(2,4-dimethylpentanenitrile) (VAZO-52).
Electronic Effects of the Group Z. Since the first
studies on RAFT polymerization, it has been reported
that by changing the nature of the Z group it is possible

3134 Benaglia et al.
Macromolecules, Vol. 38, No. 8, 2005
Table 1. Synthesized Chain-Transfer Agents and Their Precursors
Scheme 3. Synthesis of 2-Cyanoprop-2-yl
electron-donating substituents (e.g., 4-methoxy deriva-
tive 2d) reduce the control of the polymerization, as
shown by higher polydispersity values relative to those
observed with the unsubstituted derivative 2a (Table
2), with the effect being more evident in the first 2 h of
the process. The introduction of a second methoxy group
(2e) leads to a further deterioration of the performance
as reported in Table 2: PD1h: 1.40 (2d), 1.47 (2e);
PD2h: 1.28 (2d), 1.33 (2e). The 2,4-dimethoxy derivative
(2e) (PD1h 1.47) allows a better control of the polymer-
ization than the 2,4 dimethyl derivative (2c) (PD1h 1.58).
This difference is not attributed to the electron-donating
properties of the methyl group, whose inductive effect
is smaller than the mesomeric effect of methoxy group,
but is likely to reflect the bulkier nature of methyl which
leads to a reduced conjugation between the Z group and
Dithioisonicotinate
to tune the CTA characteristics in order to satisfy the
different monomer requirements2,8-11 and that all of the
different Z groups primarily influence the reactivity of
the CdS double bond toward radical addition. In this
work, we found that by varying the electronic properties
of the Z group via substitution of the phenyl with
electron-withdrawing or electron-donating substituents
or by replacing the phenyl with an electron-withdrawing
heteroaromatic residue (4-pyridinyl) the efficiency of the
CTA in MMA polymerization is affected. Precisely, the

Macromolecules, Vol. 38, No. 8, 2005
RAFT Polymerization of MMA 3135
Table 2. Number-Average Molecular Weight (Mn), Polydispersity Index (PDI), and Conversion Data for Polymerizationa
of MMA at 60 °C in the Presence of 2-Cyanoprop-2-yl Dithioesters 2 (CTA)
CTA
time (h)
Mn 10-3
PDI
% conv
CTA
time (h)
Mn 10-3
PDI
% conv
control set
1
2
275
259
315
575
1.61
1.65
2.05
2.32
13.8
21.5
37.9
99.9
2i
1
2
5.0
8.7
17
1.26
1.13
1.07
1.08
7.7
16.9
36.0
99.9
4
4
16
16
45
2a
2b
2c
2d
2e
2f
1
2
7.0
11
1.34
1.20
1.12
1.06
7.1
16.2
32.0
92.6
2j
1
2
4.3
8.8
17
1.24
1.11
1.06
1.06
6.8
16.0
31.0
99.1
4
20
4
16
55
16
59
1
2
4
32
31
34
45
1.67
1.70
1.62
1.46
9.1
16.1
31.4
97.0
2k
2l
1
2
4
5.4
10
21
59
1.20
1.11
1.06
1.07
6.6
15.0
27.2
96.2
16
16
1
2
4
9.3
13
21
49
1.58
1.42
1.22
1.12
8.2
15.0
28.4
96.1
1
2
4
4.2
7.9
15
1.24
1.13
1.07
1.06
6.8
14.8
28.7
99.9
16
16
52
1
2
4
6.8
10
18
51
1.40
1.28
1.15
1.08
7.1
14.8
30.5
94.7
2m
2n
2o
2p
1
2
4
2.6
5.8
14
1.19
1.13
1.14
1.30
5.5
9.6
23.7
59.1
16
16
39
1
2
4
9.2
10
19
53
1.47
1.33
1.20
1.08
8.0
13.8
30.6
97.7
1
2
4
6.1
10
19
53
1.29
1.17
1.09
1.07
7.2
14.9
29.6
99.8
16
16
1
2
4
6.2
10
19
54
1.31
1.22
1.11
1.06
8.2
11.8
32.3
97.3
1
2
4
6.3
11
20
55
1.30
1.19
1.10
1.07
7.7
11.5
31.5
99.8
16
16
2g
2h
1
2
4
8.7
14
26
69
1.32
1.18
1.11
1.08
7.7
15.5
30.3
97.3
1
2
4
9.1
14
23
62
1.47
1.29
1.17
1.09
8.2
17.2
32.5
99.9
16
16
1
2
4
6.9
15
28
75
1.26
1.11
1.06
1.07
5.3
14.6
31.9
98.6
16
a MMA 7.01 M, CTA 1.12 × 10-2 M, AIBN 6.1 × 10-3 M in benzene
Figure 2. SEC distribution time evolution during polymer-
ization of MMA (7.01 M) mediated by 2-cyanoprop-2-yl 2,4-
dimethyldithiobenzoate (2c) (1.12 × 10-2 M) and initiated by
AIBN (6.1 × 10-3 M) in benzene at 60 °C.
Figure 1. SEC distribution time evolution during polymer-
ization of MMA (7.01 M) mediated by 2-cyanoprop-2-yl 2,6-
dimethyldithiobenzoate (2b) (1.12 × 10-2 M) and initiated by
AIBN (6.1 × 10-3 M) at 60 °C in benzene. Each plot is scaled
relative to monomer conversions (9, 16, 31, and 97% respec-
tively).
process compared to the parent CTA 2a (Table 2). The
polymerization was also more controlled when the
phenyl was replaced with a 4-pyridinyl residue (2h) and
even more so with its salified derivative (2m) as Z group.
Thus, compounds 4-fluorophenyl (2f) (PD1h 1.31, PD2h
1.22), 2,4-difluorophenyl (2g) (PD1h 1.26, PD2h 1.13),
4-cyanophenyl (2i) (PD1h 1.26, PD2h 1.13), 4-pyridine
the dithiocarboxy moiety (see above). Conversely, the
presence of electron-withdrawing substituents (fluoro,
cyano, and trifluoromethyl) on the phenyl increases the
control of the polymerization in the early stages of the

3136 Benaglia et al.
Macromolecules, Vol. 38, No. 8, 2005
Figure 3. Molecular weight evolution with monomer conver-
sion for MMA (7.01 M) polymerization mediated by 2-cyano-
prop-2-yl dithiobenzoate (2a), 2-cyanoprop-2-yl 2,6-dimeth-
yldithiobenzoate (2b), or 2-cyanoprop-2-yl 2,4-dimethyldithio-
benzoate (2c) (1.12 × 10-2 M) and initiated by AIBN (6.1 ×
10-3 M) in benzene at 60 °C.
Figure 5. Molecular weight obtained by GPC (Mn) after 1 h
polymerization of MMA (7.01 M) mediated by different CTAs
(1.12 × 10-2 M) initiated by AIBN (6.1 × 10-3 M) at 60 °C in
benzene (empty bars) and calculated molecular weight (solid
bars) arranged by decreasing percentage difference.
consumption of the initial RAFT agent in the early
stages of the process, especially for those with electron-
donating substituents on the phenyl. The difference
between the found and the calculated Mn decreases as
the CTA efficiency increases, although the arrangement
obtained following this parameter is not perfectly
coincident with that derived from the polydispersity
values, possibly because of experimental errors affecting
the evaluation of conversion and the very close polydis-
persity values found. When the CTA is particularly
efficient, the calculated Mn is instead equal to or larger
than the observed one. This is likely due to the fact that
the experimental PMMA molecular weight values, hav-
ing been obtained by GPC calibrated with polystyrene
standards, are ca. 10% smaller than the real ones.
Electron-withdrawing Z groups, particularly the more
effective ones such as those present in compounds 2k,
2l, 2m, increase to some extent the polymerization
retardation, suggesting that the stability of the inter-
mediate radicals is slightly enhanced. Also, when the Z
group is a 4-biphenyl (2n) or a 2-naphthyl (2o), the
performance is slightly better relative to the parent
compound (2a). The higher transfer constants of the
former CTAs are reflected in their faster consumption
and smaller differences between the found and the
calculated molecular weight (Figure 5).
Solvent Effect. The solubility of 2-cyanoprop-2-yl
4-pyridiniumdithiocarboxate 4-toluenesulfonate salt (2m)
in the MMA/benzene stock solution is low at ambient
temperature but is complete at 60 °C. Although the
polydispersity of the polymers obtained in the first 2 h
is very good (Table 2), it starts to increase after the
fourth hour resulting, after 16 h, in a broad and complex
SEC trace. The polymerization is also strongly retarded.
Acetonitrile and dimethylformamide (DMF) are much
better solvents and were used to improve the CTA
solubility while performing the polymerizations at the
same molar ratios used for benzene. The SEC traces of
the polymers obtained (Figure 6) evidence that also in
acetonitrile there is a good response in the first 2 h but
after 16 h the molecular weight distribution becomes
broad and complex. On the other hand, better results
are obtained in DMF at the end of the process, although
the SEC profile of the polymer obtained after 16 h is
still complex and the polydispersity is not as low as that
of polymers obtained with other CTAs.
Figure 4. SEC traces after 1 h polymerization of MMA (7.01
M) mediated by 2-cyanoprop-2-yl dithiocarboxylate derivatives
(1.12 × 10-2 M) phenyl (2a), 2,4-dimethoxyphenyl (2e), 3,5-
bis(trifluoromethyl)phenyl (2k), and 4-pyridinium 4-toluene-
sulfonate salt (2m) and initiated by AIBN (6.1 × 10-3 M) at
60 °C in benzene. Traces normalized by height.
(2h) (PD1h 1.26, PD2h 1.11), 3,5-dicyanophenyl (2j) (PD1h
1.24, PD2h 1.11), 2,3,4,5,6-pentafluorophenyl (2l) (PD1h
1.24, PD2h 1.13), 3,5-bis(trifluoromethyl)phenyl (2k)
(PD1h 1.20, PD2h 1.11), and 4-pyridinium p-toluensul-
fonic salt (2m) (PD1h 1.19, PD2h 1.13) (Table 2) represent
a series of CTAs with increasing efficiency in MMA
polymerization, although not strictly in the above order.
As shown in Figure 4, there is a progressive narrowing
of the molecular weight distribution of the polymer after
1 h at 60 °C when switching from electron-donating to
electron-withdrawing substituents.
The addition rate of the mildly nucleophilic PMMA
propagating radical to the RAFT agent increases with
the electrophilicity of the sulfur. Thus, the transfer
constant is higher in CTAs bearing electron-withdraw-
ing Z groups. A generally larger molecular weight was
obtained in the first 2 h of polymerization compared to
the calculated values (eq 2) (Figure 5), indicating a slow
[MMA]
Mntheor
)
MWMMA‚conversion + MWCTA (2)
(
)
[CTA]

Macromolecules, Vol. 38, No. 8, 2005
RAFT Polymerization of MMA 3137
Figure 6. SEC traces of PMMA resulting from MMA (7.01 M) polymerization mediated by 2-cyanoprop-2-yl 4-pyridiniumdithio-
carboxate 4-toluenesulfonate salt (2m) (1.12 × 10-2 M) and initiated by AIBN (6.1 × 10-3 M) at 60 °C in benzene (a), acetonitrile
(b), and dimethylformamide (c).
Table 3. Number-Average Molecular Weight (Mn), Polydispersity Index (PDI), and Conversion Data for Polymerization
of MMA (7.01 M) at 60 °C in the Presence of Dithioesters 2 (CTA) (1.12 × 10-2 M) and Initiated by AIBN (6.1 × 10-3 M)
Using Benzene, Acetonitrile, and DMF as Solvent
CTA
time (h)
Mn 10-3
PDI
conv (%)
CTA
time (h)
Mn 10-3
PDI
conv (%)
2aa
1
2
7.0
11
1.34
1.20
1.12
1.06
7.1
16.2
32.0
92.6
2ma
1
2
2.6
5.8
14
1.19
1.13
1.14
1.30
5.5
9.6
23.7
59.1
4
20
4
16
55
16
39
2ab
2ac
1
2
7.0
11
1.40
1.27
1.15
1.12
10.9
22.5
38.8
99.9
2mb
2mc
1
2
3.9
10
1.12
1.07
1.08
1.21
8.7
18.4
31.0
83.1
4
19
4
20
16
52
16
52
1
2
4
9.0
14
25
64
1.39
1.28
1.13
1.07
9.3
19.1
39.3
99.9
1
2
4
2.8
6.7
21
1.26
1.14
1.09
1.14
3.5
8.8
26.8
72.4
16
16
59
a Polymerization conducted in benzene. b Polymerization conducted in acetonitrile. c Polymerization conducted in dimethylformamide.
Scheme 4. Possible Termination Reactions in RAFT Polymerization
For the sake of comparison, polymerization experi-
ments using the parent CTA 2-cyanoprop-2-yl dithioben-
zoate (2a) were also performed in acetonitrile and DMF.
These polymerizations resulted in higher conversions
but increased polydispersities of the polymers with
respect to the polymerization carried out in benzene
(Table 3). These findings can be explained by the
somewhat lower propagation rate constant (kp) for the
polymerization of MMA in benzene compared to aceto-
nitrile and DMF.29 In the absence of other effects, a
higher kp would give higher conversions but would also
reduce the chain-transfer constant (ktr/kp) of the RAFT
agent. In the case of 2-cyanoprop-2-yl 4-pyridinium-
dithiocarboxate 4-toluenesulfonate salt (2m), the situ-
ation is more complex (Table 3). While there is evidence
of a higher overall rate of polymerization in acetonitrile
and DMF when compared to benzene, there is no
indication of increased polydispersities (lower chain-
transfer constants) in the former solvents. The out-
standing features of the RAFT polymerizations with 2m
in all three solvents are rapid attainment of narrow
polydispersity, retardation of polymerization, and com-
plex molecular weight distributions at the higher con-
versions (Table 3 and Figure 6).
These observations suggest a high transfer constant
accompanied by an increase in the level of termination
reactions of the intermediate radicals 3 and 4 (Scheme
1). Indeed, DFT calculations (see below) indicate that
electron-withdrawing Z groups enhance the rate of
addition of radicals to the RAFT agent and decrease the
rate of fragmentation of the intermediate radical. These
effects would result in higher concentrations of radicals
3 and 4 and hence increase the probability of their
reaction with other radicals. The possible side reactions
for intermediate radical 4 are outlined in Scheme 4.
Aside from the termination reactions of the propagating
radicals (steps I and II), other possible termination
reactions involve combination of the intermediate radi-

3138 Benaglia et al.
Macromolecules, Vol. 38, No. 8, 2005
Table 4. Number-Average Molecular Weight (Mn), Polydispersity Index (PDI), and Conversion Data for the
Polymerizationa of MMA at 60 °C in the Presence of 2-Cyano-4-methylpent-2-yl Dithioesters (CTA)
CTA
time (h)
Mn 10-3
PDI
conv (%)
CTA
time (h)
Mn 10-3
PDI
conv (%)
2q
1
2
4.7
12
1.18
1.10
1.08
1.09
8.3
18.9
31.7
99.9
2s
1
2
2.5
7.4
16
1.13
1.06
1.04
1.06
3.9
12.8
28.2
99.9
4
24
4
16
65
16
56
2r
1
2
4.6
10
1.15
1.07
1.06
1.09
7.0
14.7
32.0
99.9
2t
1
2
3.0
7.0
16
1.24
1.13
1.06
1.06
4.4
12.0
25.4
99.9
4
21
4
16
58
16
55
a MMA 7.01 M, CTA 1.12 × 10-2 M, AIBN 6.1 × 10-3 M in benzene.
Table 5. Number-Average Molecular Weight (Mn), Polydispersity Index (PDI), and Conversion Data for Polymerizationa
of MMA at 90 °C in the Presence Dithioesters 2 (CTA) Initiated by 1,1-Azobis(cyclohexanecarbonitrile)
CTA
time (h)
Mn 10-3
PDI
conv (%)
CTA
time (h)
Mn 10-3
PDI
conv (%)
2a
1
2
9.4
16
1.28
1.15
1.11
1.09
18.2
27.6
47.3
97.6
2q
1
2
9.1
17
1.13
1.08
1.08
1.09
16.0
25.0
44.6
88.8
4
24
4
30
16
53
16
58
2i
1
2
7.3
15
1.19
1.09
1.06
1.08
15.0
25.9
45.6
95.5
2r
2s
1
2
4.2
12
1.14
1.08
1.07
1.09
8.6
18.9
32.0
83.8
4
27
4
22
16
48
16
49
2k
1
2
4
5.0
9.1
27
1.21
1.15
1.11
1.10
9.8
15.0
34.7
81.7
1
2
4
11
13
25
45
1.09
1.10
1.09
1.11
17.8
22.5
37.4
80.7
16
33
16
a MMA 6.55 M, CTA 1.13 × 10-2 M, initiator 1.8 × 10-3 M in benzene.
Scheme 5. Model of Addition and Fragmentation
Reactions Considered for DFT Calculations
cal 4 and the growing radical chain to form the three-
armed star30-33 (step III) or disproportionation to yield
a vinyl-ended chain and the reduced form of the
intermediate radical (step IV). The intermediate radical
4 can also react with 2-cyanoprop-2-yl radical derived
from the initiator decomposition (step V) (these two
radicals could also disproportionate) and dimerize to
form the four-armed star (step VI). The termination
reactions are expected to be more competitive with chain
propagation at high conversion when the monomer
concentration is lower. This could explain the greater
broadening of the molecular weight distribution toward
the end of the process.
Influence of the Leaving Group R. In the course
of this study, it was found that for all the CTAs carrying
2-cyanoprop-2-yl as the leaving group (R) and indepen-
dently from the nature of Z, it was not possible to reach
at low conversion the same narrow molecular weight
distribution obtained at higher conversion. The use of
a bulkier and better leaving group as 2-cyano-4-meth-
ylpent-2-yl confirmed that R plays a key role in the early
stages of polymerization.13 The higher stability of the
radical leaving group shifts the equilibrium from the
intermediate radical 3 (Scheme 1) to the macroCTA
(PMMA-RAFT agent) and thereby increases the chain-
transfer constant of the initial CTA. Thus, while the
group Z influences the whole process, the leaving group
R only affects the initial stages when there is more need
for improvement. The narrower molecular weight dis-
tribution of polymers obtained using 2-cyano-4-meth-
ylpent-2-yl dithiobenzoate (2q), 2-cyano-4-methylpent-
2-yl 4-cyanodithiobenzoate (2r), 2-cyano-4-methylpent-
2-yl 3,5-bistrifluoromethyldithiobenzoate (2s), or 2-cyano-
4-methylpent-2-yl 4-methoxydithiobenzoate (2t) as CTA
demonstrates a remarkably improved performance in
the first 2 h of the polymerization (Table 4). The good
leaving ability of the R group in the CTA 2-cyano-4-
methylpent-2-yl 4-methoxydithiobenzoate (2t) leads to
polymers of lower polydispersity than those obtained
with the parent compound 2a, despite the presence of
the deactivating methoxy group. Best results are ob-
tained when the favorable effect of a bulky R group is
combined with the favorable effect of electron-withdraw-
ing substituents on Z, for example, as in RAFT agent
2s (Table 4).
Effect of Temperature in the RAFT-Mediated
MMA Polymerization. Some of the more representa-
tive chain-transfer agents were used in the polymeri-
zation of MMA at 90 °C initiated by 1,1-azobis-
(cyclohexanecarbonitrile). The results (Table 5) show a
general improvement in performance in comparison to
the polymerizations conducted at 60 °C.
The higher polymerization temperature appears to
increase the propagation rate constant and the chain-
transfer constant, allowing for very narrow polidisper-
sities from the beginning of the process, and further-
more, this is achieved at lower rates of initiation.
Following the same trend for experiments at 60 °C,
electron-withdrawing substituents on the phenyl im-
proved the performance of the CTAs at 90 °C. A further
improvement was achieved using 2-cyano-4-methylpent-
2-yl as leaving group.
DFT Calculations. Enthalpy barriers for the addi-
tion and fragmentation reactions involved in the RAFT
process were calculated for the model intermediate
radicals 5 (Scheme 5) using the group CMe2COOMe (X)
to mimic the fragment derived from MMA and either

Macromolecules, Vol. 38, No. 8, 2005
RAFT Polymerization of MMA 3139
Table 6. Enthalpy Barriers for the Addition and
Fragmentation Reactions Modeling the RAFT
Polymerization of MMA in the Presence CTAs 2
Computed at the B3LYP/6-31G* Level at 333 K
6). This is in agreement with the improved polydisper-
sity in the early stages of the process reported in Table
2.
As far as the nature of the Z residue is concerned,
strong electron-withdrawing groups, such as 4-cyanophe-
nyl and 3,5-bis(trifluoromethyl)phenyl, considerably
increase the enthalpy barrier for fragmentation of S-X
and S-Y bonds and decrease the enthalpy barrier for
the addition of the X• and Y• radicals. Thus, they are
predicted to considerably enhance the addition rate of
both radicals to the dithiocarboxyl function and to
decrease the fragmentation rate of S-X and S-Y bonds.
This confirms the improved performance observed in the
first 2 h of CTAs carrying electron-withdrawing Z
groups (Table 2), which implies a faster depletion of the
initial RAFT agent.
CTA
∆H#
∆H#
∆H#
∆H#
Xadd
Xfrag
Yadd
Yfrag
2a
2q
2d
2f
5.3
10.6
7.2
8.6
5.3
5.4
5.2
4.5
4.3
10.6
10.2
10.4
11.6
10.9
7.7
7.3
7.2
6.4
6.1
6.4
8.5
8.5
9.3
8.6
2i
2k
Table 7. Radical Stabilization Energya (kcal/mol)
Computed at the B3LYP/6-31G* level
•R
BDE (kcal/mol)
•CHMeCOOMe
9.6
20.6
21.5
22.6
•CMe2COOMe (X•)
•CMe2CN (Y′•)
•CMe(CH2CHMe2)CN (Y′′•)
Conclusions
a C-H bond dissociation energy relative to the C-H BDE
The effect of substituents on the phenyl group (Z) of
dithiobenzoate RAFT agents in the polymerization of
MMA has been examined. Electron-withdrawing Z
groups, which confer a higher electrophilic character to
the thiocarboxylic sulfur, enhance the radical addition
to the CdS double bond leading to narrower polydis-
persities of the polymers from the early stages of the
polymerization. The opposite is true for electron-donat-
ing Z group. Ortho substituents on phenyl diminish the
effectiveness of the RAFT agent by reducing the degree
of conjugation of phenyl with the CdS double bond. The
beneficial steric effect of tertiary leaving groups was also
confirmed. Solvent effects were relatively minor, al-
though the lower propagation rate constant for the
polymerization of MMA in benzene (versus acetonitrile
or DMF) manifested in improved polydispersities at the
expense of conversion. An increase in the polymerization
temperature from 60 to 90 °C also proved beneficial,
apparently because of an increase in the chain-transfer
constant of the RAFT agents accompanied by the
expected higher rate of polymerization. DFT calculations
suggest that the experimental results reflect variations
of the enthalpy barrier for the addition and fragmenta-
tion reactions involved in the process with different Z
and R groups.
computed for methane.
CMe2CN (Y′) or CMe(CH2CHMe2)CN (Y′′) as leaving
group. Also, electron-donating or electron-withdrawing
moieties were used as Z group.
The calculated enthalpy barrier for the addition of the
radical X• to the CdS double bond of the CTA (∆H#
)
Xadd
is small and decreases with the increasing electron-
withdrawing ability of the Z group along the series: 3,5-
bistrifluoromethylphenyl (2k) > 4-cyanophenyl (2i) >
4-fluorophenyl (2f) > phenyl (2a) > 4-methoxyphenyl
(2d) (Table 6). Conversely, the enthalpy barrier for the
fragmentation of S-X bond (∆H#Xfrag) of radicals 5 is
higher and displays an opposite trend, that is, ∆H#
Xfrag
increases with the electron-withdrawing ability of the
Z group.
Radical 2-cyanoprop-2-yl (Y′•) is computed to be more
stable than the radical CMe2COOMe (X•) by ca. 1 kcal/
mol (Table 7). As a consequence, although similar trends
are computed for the reversible addition fragmentation
reactions of the group Y′, ∆H#
and ∆H#
are,
Yadd
Yfrag
respectively, computed to be about 2 kcal/mol smaller
and 1.5 kcal/mol larger than ∆H#Xfrag and ∆H#Xadd, and
thus, the transfer constant of CTAs 2, which affects the
polydispersity mainly in the early stages of the process
(eq 3),5 is expected to be high since the fragmentation
rate of 2-cyanoprop-2-yl (Y) is predicted to be high.
Acknowledgment. We acknowledge the valuable
contributions of several researchers from CSIRO Mo-
lecular Science: Dr. San Thang and Dr. Graeme Moad
for useful discussion and advice on the synthesis of some
of the RAFT agents, Dr. Carl Braybrook for Mass
Spectroscopy, Dr. Roger Mulder for 500 MHz NMR data,
Mr. Bill Chong and Mr. Heng Taing for some of the GPC
analyses, and Kuo-Chen Shih for the preparation of a
sample of 2-cyano-4-methylpent-2-yl dithiobenzoate.
We are also grateful to Du Pont Europe for donating
a sample of 2,2′-azobis(2,4-dimethylpentanenitrile)
(VAZO52).
kXfrag
ktr ) kXadd
(3)
kXfrag + kyfrag
The stability of the radical group is also expected to
increase with steric hindrance. Indeed, the radical
2-cyano-4-methylpent-2-yl (Y′′•) has a calculated stabi-
lization energy ca. 1 kcal/mol greater than that of
2-cyanoprop-2-yl (Y′•) and ca. 2 kcal/mol greater than
of the radical •CMe2COOMe (X•), all these three species
being considerably more stable (10 kcal/mol or more)
than the secondary radical (•CHMeCOOMe). This justi-
fies the necessity of a tertiary leaving group in CTAs
suitable for MMA polymerization (Table 7).
Thus, the transfer constant of 2-cyano-4-methylpent-
2-yl dithiobenzoate(2q) is expected to be higher than
that of 2-cyanoprop-2-yl dithiobenzoate (2a), and indeed,
the enthalpy barrier for the fragmentation of the S-Y
bond is computed to be much smaller for the intermedi-
ate radical deriving from 2q than for that from 2a (Table
References and Notes
(1) Le, T. P.; Moad G.; Rizzardo, E.; Thang, S. H. WO 98/01478
1998.
(2) Chiefari, J.; Chong, Y. K.; Ercole, F.; Krstina, 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, 5559-5562.
(3) Madayunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Krstina, J.;
Moad, G.; Postma, A. Macromolecules 2000, 33, 243-245.
(4) Rizzardo, E.; Chiefari, J.; Chong, Y. K.; Ercole, F.; Krstina,
J.; Jeffery, J.; Le T. P. T.; Mayadunne, R. T. A.; Meijs, G. F.;

3140 Benaglia et al.
Macromolecules, Vol. 38, No. 8, 2005
Moad, C. L.; Moad, G.; Thang, S. H. Macromol. Symp. 1999,
143, 291-307.
Millam, J. M.; Daniels, A. D.; Kudin, K. N.; Strain, M. C.;
Farkas, O.; Tomasi, J.; Barone, V.; Cossi, M.; Cammi, R.;
Mennucci, B.; Pomelli, C.; Adamo, C.; Clifford, S.; Ochterski,
J.; Petersson, G. A.; Ayala, P. Y.; Cui, Q.; Morokuma, K.;
Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman,
J. B.; Cioslowski, J.; Ortiz, J. V.; Stefanov, B. B.; Liu, G.;
Liashenko, A.; Piskorz, P.; Komaromi, I.; Gomperts, R.;
Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng,
C. Y.; Nanayakkara, A.; Gonzalez, C.; Challacombe, M.; Gill,
P. M. W.; Johnson, B. G.; Chen, W.; Wong, M. W.; Andres, J.
L.; Head-Gordon, M.; Replogle, E. S.; Pople, J. A. Gaussian
98, revision A.7; Gaussian, Inc.: Pittsburgh, PA, 1998.
(21) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28,
213-222.
(5) Moad, G.; Chiefari, J.; Krstina, J.; Postma, A.; Madayunne,
R. T. A.; Rizzardo, E.; Thang, S. H. Polym. Int. 2000, 49, 993-
1001.
(6) Stenzel-Rosenbaum, M.; Davis, T. P.; Chen, V.; Fane, A. G.
J. Polym. Sci., Part A: Polym. Chem. 2001, 39, 2777-2783.
(7) Quinn, J. F.; Chaplin, R. P.; Davis, T. P. J. Polym. Sci., Part
A: Polym. Chem. 2002, 40, 2956-2966.
(8) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Chong, Y.
K.; Moad, G.; Thang, S. H. Macromolecules 1999, 32, 6977-
6980.
(9) Destarac, M.; Charmot, D.; Franck, X.; Zard S. Z. Macromol.
Rapid Commun. 2000, 21, 1035-1039.
(10) Destarac, M.; Bzducha, W.; Taton, D.; Gauthier-Gillaizeau,
I.; Zard S. Z. Macromol. Rapid Commun. 2002, 23, 1049-
1054.
(11) Chiefari, J.; Mayadunne, R. T. A.; Moad, C. L.; Moad, G.;
Rizzardo, E.; Postma, A.; Skidmore, M. A.; Thang, S. H.
Macromolecules 2003, 36, 2273-2283.
(12) This mechanism has analogy in the radical chain process
reported for the photolysis of xanthates. Delduc, P.; Tailhan,
C.; Zard, S. Z. J. Chem. Soc., Chem. Commun. 1988, 308-
310.
(13) Chong, Y. K.; Krstina, J.; Le, T. P. T.; Moad, G.; Postma, A.;
Rizzardo, E.; Thang, S. H. Macromolecules 2003, 36, 2256-
2272.
(14) Perrier, S.; Takolpuckdee, P.; Westwood, J.; Lewis, D. M.
Macromolecules 2004, 37, 2709-2717.
(15) Donovan, M. S.; Lowe, A. B.; Sumerlin, B. S.; McCormick, C.
L. Macromolecules 2002, 35, 4123-4132.
(16) Barner-Kowollik, C.; Quinn, J. F.; Nguyen, T. L. U.; Heuts,
J. P. A.; Davis, T. P. Macromolecules 2001, 34, 7849-7857.
(17) Fisher, T. H.; Dershem, S. M.; Prewitt, M. L. J. Org. Chem.
1990, 55, 1040-1043.
(18) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(19) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-
789.
(20) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.;
Robb, M. A.; Cheeseman, J. R.; Zakrzewski, V. G.; Montgom-
ery, J. A., Jr.; Stratmann, R. E.; Burant, J. C.; Dapprich, S.;
(22) Scott, P. A.; Radom, L. J. Phys. Chem. 1996, 100, 16502-
16513.
(23) Thang, S. H.; Chong, Y. K.; Mayadunne, R. T. A.; Moad, G.;
Rizzardo, E. Tetrahedron Lett. 1999, 2435-2438.
(24) Bouhadir, G.; Legrand, N.; Quiclet-Sire, B.; Zard, S. Z.
Tetrahedron Lett. 1999, 277-280.
(25) Thiel, W.; Mayer, R. J. Prakt. Chem. 1989, 331, 243-262.
(26) Alberti, A.; Benaglia, M.; Laus, M.; Sparnacci, K. J. Org.
Chem. 2002, 67, 7911-7914.
(27) Zhu, J.; Zhu, X.; Cheng, Z.; Liu, F.; Lu, J. Polymer 2002, 43,
7037-7042.
(28) Zhu, J.; Zhu, X.; Zhou, D.; Chen, J.; Wang, X. Eur. Polym. J.
2004, 40, 743-749.
(29) Olaj, O. F.; Schno¨ll-Bitai, I. Monatsh. Chem. 1999, 130, 731-
740.
(30) De Brouwer, H.; Schellekens, M. A. J.; Klumperman, B.;
Monteiro, M. J.; German, A. L. J. Polym. Sci., Part A: Polym.
Chem. 2000, 38, 3596-3603.
(31) Monteiro, M. J.; de Brouwer, H. Macromolecules 2001, 34,
349-352.
(32) Kwak, Y.; Goto, A.; Tsujii, Y.; Murata, Y.; Komatsu, K.;
Fukuda, T.; Macromolecules 2002, 35, 3026-3029.
(33) Kwak, Y.; Goto, A.; Fukuda, T. Macromolecules 2004, 37,
1219-1225.
MA0480650