A multidisciplinary approach to the use of pyridinyl dithioesters and their N-oxides as CTAs in the RAFT polymerization of styrene. Not the chronicle of a failure foretold

Article abstract of DOI:10.1021/ma050652d

The efficiency of three isomeric pyridinyl dithioesters and their N-oxides as chain transfer agents in the RAFT (reversible addition fragmentation chain transfer) polymerization of styrene was tested. Ortho (dithiopicolinates), meta (ditionicotinates), an

Full text of DOI:10.1021/ma050652d

7610
Macromolecules 2005, 38, 7610-7618
A Multidisciplinary Approach to the Use of Pyridinyl Dithioesters and
Their N-Oxides as CTAs in the RAFT Polymerization of Styrene. Not
the Chronicle of a Failure Foretold
Angelo Alberti,*,† Massimo Benaglia,† Maurizio Guerra,† Mihaela Gulea,‡
Philippe Hapiot,§ Michele Laus, Dante Macciantelli,† Serge Masson,‡
Almar Postma,# and Katia Sparnacci
ISOF-CNR, Area della Ricerca, Via P. Gobetti 101, 40129 Bologna, Italy; LCMT, UMR CNRS 6507,
ENSICAEN, 6 Bd. Mare´chal Juin, 14050 Caen, France; LEMM/SESO, UMR CNRS 6510, Universite´
de Rennes 1, Institut de Chimie de Rennes, Campus de Beaulieu-Bat. 10C, 35042 Rennes, France;
Dipartimento di Scienze e Tecnologie Avanzate, INSTM, UdR Alessandria, Corso Borsalino 54,
15100 Alessandria, Italy; and CRC-for Polymers, CSIRO Molecular Science, Bag 10,
Clayton South, Victoria 3169, Australia
Received March 29, 2005; Revised Manuscript Received June 28, 2005
ABSTRACT: The efficiency of three isomeric pyridinyl dithioesters and their N-oxides as chain transfer
agents in the RAFT (reversible addition fragmentation chain transfer) polymerization of styrene was
tested. Ortho (dithiopicolinates), meta (ditionicotinates), and para (dithioisonicotinates) isomers controlled
the polymerization of styrene although with some retardation with respect to dithiobenzoates. The
retardation, which was even greater for the N-oxides, was attributed to excessive stabilization of the
dormant radical species intermediate in the RAFT process by the heteroaromatic rings as inferred from
the measured reduction potentials of the compounds. Styrene polymerization was actually blocked at
very low conversion in the case of the dithioisonicotinate N-oxide, and on the basis of ESR (electron spin
resonance) studies it is suggested that in this case the dormant radical may actually act as a scavenger
of the propagating radical. Although the knowledge beforehand of the reduction potential of a given CTA
(chain transfer agent), from which the stability of the dormant radical it would form during the RAFT
process could be estimated, might in principle allow one to foretell its performance, such predictions
must be considered with caution.
Scheme 1
Introduction
Reversible addition fragmentation chain transfer
(RAFT) polymerization based on the use of dithioesters
ZC(S)SR as chain transfer agents (CTAs) has recently
emerged as one of the most promising controlled radical
polymerization processes because of its versatility, as
it can handle the presence of a variety of different
functional monomers and it requires relatively mild
operating conditions.1-5 The key steps of the RAFT
process as shown in Scheme 1 are the thiophilic addition
of the propagating radical to the thiocarbonyl group of
the dithioester and the fragmentation of the sulfur-
carbon bond of the resulting spin adduct to restitute a
dithioester and a propagating radical.
For the RAFT process to be efficient, the R residue of
the CTA must be a good leaving group, e.g. 2-cyanoprop-
2-yl, cumyl, benzyl;6 besides, the spin adduct 1 must be
a relatively stable radical, its formation, i.e. the addition
of the propagating radical to the dithioester, being
competitive with propagation.7,8 On the other hand, an
excessive stability of 1 would result in a slow fragmen-
tation reaction and hence in an undesired retardation
of the polymerization.9-12 The stability of the spin
adduct 1 is therefore critical for the efficiency of the
RAFT process and can be modulated by changing the
nature of the residue Z. Indeed, electron-withdrawing
Z groups make radicals 1 longer-lived due to the capto-
dative effect, i.e. stabilization due to the simultaneous
presence of electron-donating and electron-accepting
substituents bound to the radical center,13,14 whereas
the radical stabilization effect of lone pair donors Z
groups is reduced. It was recently demonstrated that,
in RAFT radicals, Z groups that are strong lone pair
donors and weak sigma acceptors (such as -NR2)
remain as neat radical stabilizing substituents, but
those that are weaker lone pair donors and stronger
sigma acceptors (such as -OR) have a negligible radical
stabilization effect and (in some cases) even a destabi-
lizing effect, due to their sigma withdrawing proper-
ties.15 A Z phenyl group is sufficient to stabilize radicals
1, so that tertiary dithiobenzoates are good CTAs in the
polymerization of MMA or styrene.1,2 A similar, actually
slightly stronger stabilizing effect is exerted by the
phosphoryl group of phosphoryl dithioformates [(EtO)2P-
(O)C(S)SR] that have also been successfully exploited
as controlling agents in the polymerization of styrene.16
In addition, phosphoryl dithioformates proved particu-
larly useful for ESR studies of the polymerization
process.17
Following the recent report of the synthesis of the pre-
viously unknown N-oxides of some pyridinyl dithioes-
ters,18 we were prompted to investigate whether the
electron-withdrawing effect of the heteroaromatic ring
might render these compounds susceptible of being good
CTAs. We report here on the polymerization controlling
* Corresponding author: e-mail aalberti@isof.cnr.it.
† ISOF.
‡ ENSICAEN.
§ LEMM/SESO.
INSTM.
# CRC-for Polymers, CSIRO.
10.1021/ma050652d CCC: $30.25 © 2005 American Chemical Society
Published on Web 08/12/2005

Macromolecules, Vol. 38, No. 18, 2005
Using Pyridinyl Dithioesters as RAFT CT Agents 7611
CHPy), 8.33 (d, 1H, J ) 8.1, CHPy), 8.59 (dt, 1H, J ) 4.6, J )
0.7, CHPy). 13C NMR (CDCl3): δ ) 41.6 (SCH2), 122.2, 126.8,
127.6, 2 × 128.6, 2 × 129.4, 135.0, 136.9, 147.9, 156.4 (C6H5
+ C5H4N), 226.1 (CdS). HRMS: (MH+) calcd 246.0411, found
246.0393.
ability of benzyl 2-dithiopicolinate (2), benzyl dithioni-
cotinate (3), benzyl dithioisonicotinate (4), and of their
corresponding N-oxides (5-7). We also report on the
redox properties of compounds 2-7, of methyl 2-dithi-
opicolinate (8), and of methyl dithioisonicotinate N-oxide
(9) that have been investigated for the sake of compari-
son as well as on the ESR characterization of their
radical anions and of some spin adducts modeling the
radical species involved in the RAFT polymerization
process.
Benzyl Dithionicotinate, 3. Pink solid, mp 29 °C. Yield
91%. 1H NMR (CDCl3): δ ) 4.67 (s, 2H, SCH2), 7.08-7.67 (m,
6H, C6H5 + CHPy), 8.22 (dt, 1H, J ) 8.0, J ) 2.0, CHPy), 8.73
(dd, 1H, J ) 4.8, J ) 1.4, CHPy), 9.16 (d, 1H, J ) 2.1, CHPy).
13C NMR (CDCl3): δ ) 42.4 (SCH2), 123.1, 127.9, 2 × 128.2, 2
× 129.3, 134.2, 134.4, 140.1, 147.0, 152.7 (C6H5 + C5H4N),
224.1 (CdS). HRMS: (MH+) calcd 246.0411, found 246.0400.
Benzyl Dithioisonicotinate, 4. Red paste. Yield 89%. 1H
NMR (CDCl3): δ ) 4.56 (s, 2H, SCH2), 7.18-7.34 (m, 5H,
C6H5), 7.70 (d, 2H, J ) 6.1, CHPy), 8.65 (d, 2H, J ) 6.1, CHPy).
13C NMR (CDCl3): δ ) 42.3 (SCH2), 2 × 120.1, 128.0, 2 ×
128.8, 2 × 129.3, 134.2, 150.0, 2 × 150.4, (C6H5 + C5H4N),
224.9 (CdS). HRMS: (MH+) calcd 246.0411, found 246.0396.
Benzyl 2-Dithiopicolinate N-Oxide, 5. Potassium tert-
butoxide (3.36 g, 30 mmol) was added under stirring to a
mixture of 2-benzenesulfonylmethylpyridine N-oxide22 (2.49 g,
10 mmol) and elemental sulfur (0.96 g, 30 mmol) in THF (100
mL). During the addition the color of the mixture changed to
dark brown. After stirring the reaction mixture up to 12 h
benzyl bromide (5.13 g, 30 mmol) was added dropwise, and
stirring continued for 1 h. The solvent was then removed under
reduced pressure, and the residue was dissolved in methylene
chloride (10 mL) and chromatographed on silica gel. Upon
removal of the solvent, the pure dithioester 5 was isolated.
Dark red paste. Yield 84%. 1H NMR (CDCl3): δ ) 4.51 (s, 2H,
SCH2), 7.15-7.40 (m, 7H, C6H5 + 2 × CHPy), 8.08 (m, 1H,
CHPy), 8.21 (m, 1H, CHPy). 13C NMR (CDCl3): δ ) 43.9 (SCH2),
125.6, 126.9, 128.1, 128.5, 2 × 129.1, 2 × 129.9, 134.6, 140.5,
149.4 (C6H5 + C5H4N), 216.8 (CdS). HRMS: (MH+) calcd
262.0360, found 262.0349.
Experimental Section
Materials and Methods. R,R-Azoisobutyronitrile (Fluka,
98%), benzyl bromide (Aldrich, 98%), bromomethane (Aldrich,
99.5%), copper(I) bromide (Aldrich), dimethylmercury (Aldrich,
95%), 4,4′-dinonyl-2,2′-dipyridyl (Aldrich, 97%), di-tert-butyl
peroxide (BPO, Fluka, 95%), manganese(0) carbonyl (Aldrich,
98%), tetrabutylammonium perchlorate (Fluka, >99%), tet-
rabutylammonium tetrafluoroborate (Fluka, >99%), triphen-
ylgermanium hydride (Aldrich), and tris(trimethylsilyl)silane
(Aldrich, 97%) were commercially available. Styrene (99%) was
purchased from Aldrich and washed with 3 × 100 mL of 2.0
M sodium hydroxide and 3 × 100 mL of water, dried with
anhydrous sodium sulfate, stored at 5 °C, and eventually
distilled under vacuum prior to use. Compounds 8 and 9 were
prepared as previously described.18 Acetonitrile (ACN), di-
methyl sulfoxide (DMSO), hexamethylphosphoramide (HMPA),
tetrahydrofuran (THF), and all other solvents (Aldrich) were
dried and distilled as necessary.
The reactions were monitored by TLC (thin layer chroma-
tography) using silica plates. The products were purified by
flash chromatography and crystallized when needed. NMR
spectra were recorded with a Bruker DPX250 spectrometer
(1H, 250 MHz; 13C, 62.9 MHz) using tetramethylsilane (TMS)
as internal standard. Chemical shifts (δ) are given in ppm and
coupling constants (J) in Hz. High-resolution mass spectra
were recorded with a QTOF Micro Waters spectrometer in the
positive-ion electrospray-ionization mode.
Benzyl Dithionicotinate N-Oxide, 6. Dark pink solid, mp
1
64 °C. Yield 95%. H NMR (CDCl3): δ ) 4.59 (s, 2H, SCH2),
7.22-7.35 (m, 6H, C6H5 + CHPy), 7.78 (d, 1H, J ) 8.1, CHPy),
8.30 (d, 1H, J ) 6.4, CHPy), 8.77 (s, 1H, CHPy). 13C NMR
(CDCl3): δ ) 42.4 (SCH2), 123.6, 125.4, 126.9, 128.2, 2 × 128.9,
2 × 129.3, 133.8, 137.2, 141.0, 142.8 (C6H5 + C5H4N), 219.5
(CdS). HRMS: (MH+) calcd 262.0360, found 262.0341.
Benzyl Dithioisonicotinate N-Oxide, 7. Dark red solid,
mp 57 °C. Yield 74%. 1H NMR (CDCl3): δ ) 4.59 (s, 2H, SCH2),
7.18-7.38 (m, 5H, C6H5), 7.93 (d, 2H, J ) 7.4, 2 × CHPy), 8.13
(d, 2H, J ) 7.4, 2 × CHPy). 13C NMR (CDCl3): δ ) 42.3 (SCH2),
2 × 123.3, 128.1, 2 × 128.8, 2 × 129.3, 2 × 134.1, 138.8, 138.9
(C6H5 + C5H4N), 219.3 (CdS). HRMS: (MNa+) calcd 284.0180,
found 284.0161.
Synthesis of Polystyryl Bromide. A solution comprising
styrene (5.68 mL, 49.0 mmol), copper(I) bromide (70.6 mg, 0.53
mmol), 4,4′-dinonyl-2,2′-dipyridyl (403.6 mg, 1 mmol), and
1-phenylethyl bromide (0.34 mL, 2.5 mmol) was prepared and
transferred to an ampule that was subsequently degassed by
three freeze-evacuate-thaw cycles, sealed, and heated at 110
°C for 7 h in a thermostated oil bath. The conversion was
estimated to be 67.7% through a comparison of the integrals
of the NMR doublets centered at 5.35 and 5.75 ppm (2H,
PhCHdCH2) and at 6.4-7.4 ppm (5H, PhCHdCH2). The
solution was diluted in chloroform, precipitated in methanol,
and filtered. Gel permeation chromatography (GPC) analysis
gave Mn 1205 and Mw/Mn 1.1.
Synthesis of Dithioesters. Compounds 2-7 were synthe-
sized according to a general procedure whereby potassium tert-
butoxide was added to a mixture of the appropriate benzene-
sulfonylmethylpyridine19-21 or benzenesulfonylmethylpyridine
N-oxide22 and elemental sulfur in THF. The resulting reaction
mixture was further reacted with benzyl bromide and eventu-
ally chromatographed on silica gel. Upon removal of the
solvent, the pure dithioesters 2-7 were isolated.
Benzyl 2-Dithiopicolinate, 2. Potassium tert-butoxide
(3.36 g, 30 mmol) was added under stirring to a mixture of
2-benzenesulfonylmethylpyridine (2.33 g, 10 mmol) and el-
emental sulfur (0.96 g, 30 mmol) in THF (100 mL). During
the addition the color of the mixture changed to dark brown.
After stirring the reaction mixture up to 12 h benzyl bromide
(5.13 g, 30 mmol) was added dropwise, and stirring continued
for 1 h. The solvent was then removed under reduced pressure;
the residue was dissolved in methylene chloride (10 mL) and
chromatographed on silica gel. Upon removal of the solvent,
the pure dithioester 2 was isolated. Dark red solid, mp 35 °C.
Electrochemistry. All the cyclic voltammetry experiments
were carried out at 20 ( 1 °C in ACN using tetrabutylammo-
nium tetrafluoroborate as supporting electrolyte in a water
thermostated cell. The working electrode was either a gold-
platinum or a glassy carbon disk (φ ) 1 mm) and was carefully
polished before each set of voltammograms with a 1 µm
diamond paste and ultrasonically rinsed in absolute ethanol.
Similar electrochemical patterns were obtained in either case,
indicating that the reduction processes were not considerably
dependent on the nature of the electrode. The electrochemical
instrumentation consisted of a Tacussel GSTP4 programmer
and a home-built potentiostat equipped with a positive feed-
back compensative device.23 The data were acquired with a
1
Yield 88%. H NMR (CDCl3): δ ) 4.54 (s, 2H, SCH2), 7.18-
7.47 (m, 6H, C6H5 + CHPy), 7.78 (td, 1H, J ) 9.5, J ) 1.6,

7612 Alberti et al.
Macromolecules, Vol. 38, No. 18, 2005
Table 1. Formal Potentials (E°) of 2-9 and Some
Reference Compounds in Acetonitrile vs SCE (Saturated
Calomel electrode)
thaw cycles, sealed under nitrogen, and heated at the ap-
propriate temperature. At the end of the reaction, each ampule
was quenched in cold water and the reaction mixture diluted
with methylene chloride. The polymer was then precipitated
into methanol, washed with methanol, and purified by pre-
cipitation from methylene chloride into methanol. The polymer
was dried on silica gel in vacuo for several hours. Conversion
of styrene was estimated by weighting the obtained polymer.
Several polystyrene samples were synthesized this way per-
forming the polymerizations at either 60 or 80 °C for different
reaction times. As a typical example, 3.0 mL (26 mmol) of
styrene were reacted with 6.84 mg (28 µmol) of 4 and 0.85 mg
(5.1 µmol) of AIBN at 60 °C for 30 h, giving sample 4/2 (see
Table 4) with a yield of 7.6%. Number-average molar mass
and polydispersity index resulted in Mn ) 17 300 and Mw/Mn
) 1.37. As a further example, 3.0 mL (26 mmol) of styrene
was reacted with 7.32 mg (28 µmol) of 7 and 0.85 mg (5.1 µmol)
of AIBN at 60 °C for 30 h, giving sample 7/1 (see Table 5) with
a yield of 1.0%. Number-average molar mass and polydisper-
sity index resulted in Mn ) 5900 and Mw/Mn ) 1.21.
DFT (Density Functional Theory) Calculations. DFT
calculations employing the B3LYP functional25,26 were carried
out to compute the hyperfine splitting (hfs) constants of the
radical species 4•-, 9•-, 11, 21, and 22 using the GAUSSIAN98
system of programs.27 Unrestricted wave functions were used
to describe radical species. Geometries and hfs constants were
obtained employing a valence double-ú basis set supplemented
with standard polarization28,29 and diffuse29,30 functions on
heavy atoms (6-31+G*). At this level of theory the radical
anions 4•- and 9•- were computed to be thermodynamically
stable with respect to electron loss while the radical dianions
21 and 22 were computed to be not only thermodynamically
but also kinetically unstable. However, addition of standard
diffuse functions to heavy atoms to better describe the
negatively charged species provides reliable results in this
particular case (delocalized π dianion radicals) since inspection
of the electronic distribution in the singly occupied MO shows
that the diffuse atomic functions are less populated than the
valence atomic functions. That is, the wave function does not
describe an anion interacting with a free electron.31
compd
E°/V
lifetime/s
compd
E°/V
ref
2
3
4
-1.140 0.1-0.3
-1.150 0.4-0.7
-1.018 >20
MeC(S)SMe
∼-1.65 32
-1.64 33
(EtO)2P(O)CH2C(S)-
SMe
5
6
7
8
9
-1.008 0.1-0.2
-1.005 0.1-0.2
-0.927 10-20
-1.186 5-10
-0.987 >20
PhC(S)SMe
-1.32 34
-1.10 35
-1.04 35
-0.88 36
(EtO)2P(O)C(S)SMe
(EtO)2P(S)C(S)SMe
MeO(O)CC(S)SMe
310 Nicolet oscilloscope. The counter electrode was a Pt wire
and the reference electrode an aqueous saturated calomel
electrode (SCE) with a salt bridge containing the supporting
electrolyte. The SCE was checked against the ferrocene/
ferricinium couple (considering a formal potential E° ) +0.45
V/SCE in ACN) before and after each experiment. On the basis
of repetitive measurements, absolute errors on potentials were
found to be ca. (10 mV.
ESR Experiments. ESR spectra were recorded with an
upgraded Bruker ER200D/ESP300 spectrometer equipped
with a dedicated data station for the acquisition and manipu-
lation of the spectra, a standard variable temperature device,
a NMR gaussmeter for the calibration of the magnetic field,
and a frequency counter for the determination of g-factors that
were corrected with respect to that of perylene radical cation
in concentrated sulfuric acid. Computer simulations of the
spectra were obtained using a software24 based on a Monte
Carlo minimization procedure.
The radical anions from 2-9 were obtained either by
reduction of the appropriate compound with potassium tert-
butoxide (tBuOK) in DMSO or by electrochemical reduction.
This was carried out using an Amel Instruments 2051 poten-
tiostat and a flat cell (50 × 9.5 × 1.5 mm) inserted inside the
cavity of the ESR spectrometer and equipped with a platinum
gauze (cathode) and a platinum wire (anode). In these experi-
ments the dithioesters (ca. 10-2 M) were dissolved in dry
DMSO or ACN containing nBu4NClO4 (10-1 M) as supporting
electrolyte.
Results
In a typical experiment of radical addition, a Suprasil quartz
tube (i.d. 4 mm) containing a thoroughly argon purged benzene
or tert-butylbenzene solution of dithioester 7 or 9 (ca. 10-3 M),
and the appropriate radical precursors were irradiated with
the light from a 1 kW high-pressure mercury lamp inside the
cavity of the ESR spectrometer and, when necessary, heated.
Methyl and benzyl radicals were obtained by thermolysis or
photolysis of dimethylmercury or by photolysis of methyl or
benzyl iodide in the presence of manganese(0) carbonyl. The
polystyryl radical was similarly obtained by photolysis of the
corresponding bromide in the presence of manganese(0) car-
bonyl. Silyl and germyl radicals where instead generated by
photolysis of the corresponding organometallic hydride in the
presence of di-tert-butyl peroxide.
Electrochemical Reduction. The electrochemical
behavior of compounds 2-9 was investigated by cyclic
voltammetry (CV). The reduction of 4, 7, 8, and 9 was
a monoelectronic process reversible at any potential
scan rate, resulting in the formation of the radical
anions 4•-, 7•-, 8•-, and 9•-. As for the other derivatives,
the reduction process, also monoelectronic, depended on
the scan rate (υ) being reversible at υ > 1 V s-1 (2 and
6), υ > 2 V s-1 (5), and υ > 10 V s-1 (3). From these
scan rates the lifetime of the different radical anions
could be estimated (see Table 1). These are very
persistent species, with lifetimes ranging from some
tenths of a second for ortho and meta derivatives to over
10 s for the para isomers, their chemical stability not
being affected when replacing a methyl for a benzyl
group or when switching from pyridine derivatives to
the corresponding N-oxides. The reduction potentials
RAFT Polymerization of Styrene. A master batch of 20
mL (175 mmol) of styrene, 5.6 mg (34.1 µmol) of AIBN, and
45.6 mg (0.186 mmol) of 2-4 or 48.8 mg (0.186 mmol) of 5-7
was prepared, and aliquots of 3 mL were placed in polymer-
ization ampules. The ampules were degassed by freeze and
Table 2. ESR Hyperfine Splitting Constants (a) and g Factors (g) for Radical Anions Obtained by Reduction of
Compounds 2-9a
compd
solvent
acetonitrile
dimethyl sulfoxide
acetonitrile
a2
a3
0.101
0.124 (1N)
0.048
0.107
0.172 (1N)
0.115
0.089
0.107
a4
a5
a6
aX
g
2
3
4
5
6
7
8
9
0.148 (1N)
0.413
0.305
0.394 (1N)
0.398
0.335
0.673
0.580
0.272 (1N)
0.883
0.620
0.496 (1N)
0.670
0.513 (1N)
0.054
0.099
0.048
0.141
0.070
0.115
0.050
0.107
0.234
0.282
0.305
0.455
0.269
0.335
0.233
0.329
0.093 (2H)
n.r.b
0.044 (2H)
0.315 (1H + 1H)
n.r.b
0.055 (2H)
0.154 (3H)
0.114 (3H)
2.00763
2.00749
2.00811
2.00743
2.00758
2.00791
2.00793
2.00811
hexamethylphosphoramide
dimethyl sulfoxide
dimethyl sulfoxide
acetonitrile
0.140 (1N)
0.329
dimethyl sulfoxide
a Hyperfine splitting constants in mT. b Not resolved.

Macromolecules, Vol. 38, No. 18, 2005
Using Pyridinyl Dithioesters as RAFT CT Agents 7613
Table 3. ESR Hyperfine Splitting Constants (a) and g Factors (g) for Radical Adducts to Compounds 7 and 9 at Room
Temperature in Benzenea
compound/radical
spin adduct
a2,6
a3,5
a4
aX
aY
g
7/Me
10
11
12
13
14
15
16
17
18
19
0.385
0.388
0.398
0.375
0.384
0.384
0.384
0.396
0.392
0.386
0.180
0.177
0.178
0.170
0.179
0.168
0.179
0.173
0.180
0.173
0.394 (1N)
0.395 (1N)
0.408 (1N)
0.400 (1N)
0.395 (1N)
0.402 (1N)
0.397 (1N)
0.402 (1N)
0.395 (1N)
0.395 (1N)
0.112 (2H)
0.033 (2H)
n.r.b
0.109 (3H)
0.033 (2H)
2.00524
2.00499
2.00532
2.00529
2.00509
2.00503
2.00525
2.00416
2.00417
2.00479
7/CH2Ph
7/Si(SiMe3)3
7/GePh3
n.r.
7/polystyryl
9/Me
0.054 (2H)
0.089 (3H)
0.113 (3H)
0.094 (3H)
0.129 (3H)
e0.040 (3H)
n.r.
0.089 (3H)
0.112 (2H)
9/CH2Ph
9/Si(SiMe3)3
9/GePh3
9/polystyryl
a Hyperfine splitting constants in mT. b Not resolved.
Table 4. Molar Mass, Polydispersity Index, and
Table 5. Molar Mass, Polydispersity Index, and
Conversion Data for Polystyrene Samples Prepared via
AIBN-Initiated Polymerization of Styrene with
Dithioesters 5-7
Conversion Data for Polystyrene Samples Prepared via
AIBN-Initiated Polymerization of Styrene with
Dithioesters 2-4
dithioester/
sample
dithioester/
sample
T/°C time/h Mn × 10-3 Mw × 10-3 Mw/Mn conv/%
T/°C time/h Mn × 10-3 Mw × 10-3 Mw/Mn conv/%
2/1
2/2
2/3
2/4
2/5
2/6
2/7
2/8
2/9
3/1
3/2
3/3
3/4
3/5
3/6
3/7
3/8
3/9
3/10
4/1
4/2
4/3
4/4
4/5
4/6
4/7
4/8
4/9
60
60
60
60
80
80
80
80
80
60
60
60
60
80
80
80
80
80
80
60
60
60
60
80
80
80
80
80
14
20
30
40
4
15.0
18.4
25.7
36.5
20.8
21.4
28.4
32.9
34.0
9.5
14.1
24.3
32.3
13.7
20.1
23.2
30.3
36.6
39.2
6.3
17.3
25.3
41.8
18.9
19.8
22.9
24.8
30.1
17.6
23.1
34.6
49.5
26.7
27.9
39.6
50.1
49.1
11.1
16.5
33.7
49.3
16.5
24.1
28.3
39.9
49.4
64.1
8.1
23.8
37.4
69.8
22.7
25.8
31.8
35.2
42.1
1.17
1.25
1.34
1.35
1.28
1.30
1.39
1.50
1.40
1.15
1.17
1.38
1.52
1.20
1.19
1.21
1.31
1.35
1.63
1.27
1.37
1.40
1.60
1.20
1.30
1.39
1.42
1.4
6.9
5/1
5/2
5/3
5/4
5/5
5/6
5/7
5/8
5/9
5/10
6/1
6/2
6/3
6/4
6/5
6/6
6/7
6/8
6/9
7/1
7/2
7/3
7/4
7/5
7/6
7/7
7/8
7/9
7/10
60
60
60
60
80
80
80
80
80
80
60
60
60
60
80
80
80
80
80
60
60
60
60
80
80
80
80
80
80
20
30
41
60
4
8.2
16.0
22.9
30.0
10.5
17.0
20.5
37.8
33.6
41.4
15.9
20.1
28.5
72.3
6.3
13.0
24.0
41.4
58.0
22.2
33.0
38.8
51.1
76.8
86.0
21.6
31.5
48.5
125.0
8.0
16.2
19.3
43.1
98.7
7.0
12.6
21.7
70.0
6.0
1.58
1.50
1.80
1.90
2.00
1.90
1.89
1.80
2.20
2.00
1.35
1.56
1.70
1.72
1.28
1.37
1.42
1.67
2.20
1.21
1.30
1.40
1.64
1.27
1.24
1.42
1.50
1.56
2.00
1.5
7.8
4.3
14.1
16.0
11.7
14.8
22.4
20.9
24.4
6.4
6.9
11.5
5.0
8
8
7.0
14
20
30
14
20
30
40
4
14
20
30
40
30
40
65
105
4
11.3
13.0
17.4
18.6
4.8
10.6
14.2
23.3
10.5
17.2
20.6
24.8
28.9
32.5
1.0
5.7
9.0
19.9
1.0
6.0
7.9
8
14
20
30
40
14
30
40
65
4
8
11.8
13.5
25.8
44.5
5.9
14
20
40
30
40
65
105
4
10.9
22.7
1.0
7.6
9.6
1.5
10.6
14.3
7.8
15.0
42.6
4.6
2.2
9.0
1.1
8
8.7
8
9.1
11.3
17.3
33.0
44.3
81.6
1.5
14
20
30
10.7
13.4
15.6
14
20
30
40
12.1
21.9
28.3
33.7
5.2
7.3
10.5
20.6
determined for the different compounds are collected in
Table 1 along with those reported in the literature for
some reference derivatives. Although for some of the
compounds additional reduction waves were detected at
more negative potentials, the full electrochemical be-
havior of compounds 2-9 was not investigated as only
their first reduction potential values (that is, the
potentials corresponding to the formation of their radical
anions) were pertinent to the present study.
collected in Table 1, spectra were recorded (see Figure
1) that were attributed to the corresponding radical
The reduction potentials proved only slightly sensitive
to the relative position of the dithioester group and the
nitrogen atom, the para isomers exhibiting somewhat
lower values than the ortho and meta derivatives. Also,
replacing a methyl for a benzyl group as R residue (e.g.
switching from 2 to 8 or from 7 to 9) resulted in an
almost unnoticeable increase of the reduction potential
which, all in all, seems to be mostly dictated by the
dithioester function. According to sensible expectations,
the reduction potentials of N-oxides 5-7 were slightly
less negative than those of the corresponding com-
pounds 2-4.
Figure 1. ESR spectrum observed by electrochemical reduc-
tion at room temperature of compound 9 in dimethyl sulfoxide
n
containing Bu4ClO4 (0.1 M) as supporting electrolyte.
anions, the ESR spectral parameters of which are
collected in Table 2. The chemical reduction of the
compounds (treatment with tBuOK in DMSO or HMPA)
led to the detection of the same ESR spectra.
An increase of the applied potential during the
reduction of compounds 2 and 4 and of the correspond-
ESR Studies. When compounds 2-9 were reduced
electrochemically at room temperature inside the cavity
of the ESR spectrometer at potentials close to those

7614 Alberti et al.
Macromolecules, Vol. 38, No. 18, 2005
Figure 2. ESR spectra observed upon prolonged electrochemical reduction of compound 2 or 5 (a, radical 20), 4 (b, radical 21),
and 7 (c, radicals 21 and 22 in a 45.3% and 54.7% amount, respectively; blue, experimental; red, simulation).
Scheme 2
RAFT Polymerization of Styrene. Several poly-
styrene samples were synthesized by performing RAFT-
controlled polymerizations at either 60 or 80 °C, as
described in the Experimental Section. Number-average
molar mass and polydispersity index values are collected
in Tables 4 and 5 along with reaction yields. The
resulting polymer samples exhibited a pale red-pinkish
color due to the presence of the terminal dithioester
group of the controlling agent. In all cases size exclusion
chromatography (SEC) curves were shifted toward
higher values along the molar mass scale with increas-
ing reaction time, this increase in the molar mass being
the expected trend for a controlled polymerization
process.
ing N-oxides 5 and 7 led to the observation of new
signals that eventually replaced the initially observed
spectra of the radical anions. In the case of 2 and 5, the
same spectrum was eventually observed which exhibited
the following spectral parameters: aH ) 0.08 mT, aH
)
An increase of the polydispersity index value with
time was also common to all samples. This behavior is
similar to that already observed in the case of benzyl
diethoxyphosphoryldithioformate15 and collides with
that exhibited by benzyl dithiobenzoate,1,17 the latter
compound being a better RAFT controlling agent than
the former.
0.230 mT, aN ) 0.233 mT, aH ) 0.455 mT, g ) 2.00800
(see Figure 2a). In the case of compound 4 the spectrum
of the radical anion was replaced by the spectrum shown
in Figure 2b, which is due to a radical species where
the unpaired electron is coupled with two sets of
equivalent hydrogen atoms and a nitrogen (aN ) 0.335
mT, a2H ) 0.013 mT, a2H ) 0.272 mT, g ) 2.00816).
Similar treatment of 7 led instead to the observation of
the spectrum in Figure 2c that was in fact the super-
imposition of the signals from two radicals. One was the
same observed with dithioester 4, whereas in the other
the unpaired electron was again coupled with two sets
of equivalent hydrogen atom and a nitrogen, but with
different coupling constant (aN ) 0.601 mT, a2H ) 0.027
mT, a2H ) 0.268 mT, g ) 2.00816).
As outlined in Scheme 2, compounds 7 and 9 were
also subjected to addition by free radicals generated in
situ. Thus, 7 and 9 were thermally or photoreacted with
dimethylmercury, methyl, benzyl, or polystyryl bromide
and manganese(0) carbonyl, tris(trimethylsilyl)silane or
triphenylgermanium hydride, and BPO. The spin ad-
ducts of the in situ generated free radicals were char-
acterized by ESR spectroscopy, and their spectral
parameters are collected in Table 3.
Discussion
We have already pointed out in the introductory
section that the relative stability of the radicals of
general structure 1 is a critical factor in determining
the efficiency of a RAFT process involving a particular
chain transfer agent. Because radicals 1 have two
electron-donating thioalkyl groups linked to the radical
carbon center, their stability is bound to increase with
the electron-withdrawing ability of the third residue Z
due to capto-dative stabilization.
In this light, it should be possible to predict the ability
of a given dithioester ZC(S)SR to act as a RAFT
mediator on the basis of its reduction potential, which
is expected to lower as the electron-withdrawing ability
of Z increases.

Macromolecules, Vol. 38, No. 18, 2005
Using Pyridinyl Dithioesters as RAFT CT Agents 7615
Scheme 3
Figure 3. Experimental and DFT (density functional theory)
calculated (italic) hyperfine splitting constants for radical
dianions 20-22.
From the data collected in Table 1 it emerges that
compounds 2-9 have reduction potentials in the range
-0.9 to -1.2 V vs SCE, thus indicating that the electron-
withdrawing ability of the pyridinyl residues and of
their N-oxides is in between that of a phenyl group and
that of a methoxycarbonyl group and is comparable to
that of a phosphoryl or thiophosphoryl residue. Indeed,
thanks to this stabilizing effect, the ESR spectra of the
radical anions from compounds 2-9 could be readily
observed (see Figure 1), as it had already been the case
for the radical anions of several phosphoryl dithiofor-
mates.35 It should also be noticed that the radical anions
2•--9•- were observed independently of the relative
position of the heterocyclic nitrogen and the dithioester
function: this would indicate that direct conjugation
between the thiocarbonyl group an the nitrogen atom
is not essential to stabilization, which is not unexpected
as the radical anions of simple dithiobenzoates have also
been long since detected by ESR.
Although similar and less negative reduction poten-
tials would be expected for the ortho and para deriva-
tives with respect to the meta isomers, the values
reported in Table 1 indicate that in fact ortho and meta
derivatives exhibit similar values that are slightly more
negative than those of the corresponding para isomers.
This finding may reflect some steric interaction that
would push the dithioester group out of the plane of the
aromatic ring in the ortho derivatives, thus reducing
conjugation with respect to the para isomers.
DFT calculations at UB3LYP/6-31+G* level were
performed for the radical anions from two of the
examined dithioestersand the predicted hfs constants
(a2 ) -0.295 mT, a3 ) 0.065 mT, aN ) 0.261 mT, a5 )
0.055 mT, a6 ) -0.272 mT, a2H ) -0.016 mT for 4•-
and a2 ) -0.370 mT, a3 ) 0.172 mT, aN ) 0.472 mT, a5
) 0.192 mT, a6 ) -0.372 mT, a3H ) 0.104 mT for 9•-)
were found to be in very good agreement with the
experimentally measured values.
As for the identification of the species responsible for
the spectra observed upon prolonged treatment of 4 and
7, we believe these species to be the radical dianions
21 and 22 resulting by loss of a benzyl fragment from
the corresponding radical anions followed by further
reduction in situ (see Figure 3). This assignment was
substantiated by DFT calculations at the UB3LYP/
6-31+G* level that predicted for 21 and 22 coupling
constants in very good agreement with those found
experimentally.
detected in the chemical or electrochemical reduction
of diethoxyphosphoryl dithioformates (EtO)2P(O)C(S)-
SR.35
The stabilizing action of the heteroaromatic rings
should make the spin adducts resulting from radical
addition to dithioesters 2-9 persistent enough to readily
allow their ESR detection. In the present study radical
addition experiments were only carried out with com-
pounds 7 and 9 that, in view of their symmetric
structure, were expected to afford spin adducts with
simple ESR spectra. Thus, generation of methyl, benzyl,
silyl, germyl, and polystyryl radicals in the presence of
these compounds led to the ESR detection of the
corresponding spin adducts 10-19, for which a mesom-
erism can be viewed between structures A and B (see
Scheme 3).
Also in this case, DFT calculations predicted hfs
constants (a2,6 ) -0.421 mT, a3,5 ) 0.288 mT, aN ) 0.305
mT, aX,Y(2H) ) 0.069 mT for 11) in substantial agree-
ment with experiment. Calculations also indicated a
significant contribution of structure B, formally a divi-
nyl nitroxide, to the mesomeric system. It should how-
ever be pointed out that, although the nitrogen splitting
values for radicals 10-19 do meet expected values for
unsaturated nitroxides,37 their g-factors slightly lower
than expected also indicate that the contribution of
mesomeric structure A cannot be disregarded.
In view of their “favorable” reduction potentials and
hence of the relative stability of their spin adducts,
compounds 2-7 were tested as CTAs in the RAFT
polymerization of styrene. Compounds 8 and 9 were
instead disregarded because the methyl residue is not
a suitable leaving group. The results (see Tables 4 and
5) indicated control of the polymerization of styrene by
all the examined compounds, although with different
performances.
Figure 4 shows the trend of the number-average
molar mass Mn and the polydispersity index Mw/Mn as
a function of time at 60 °C for the polystyrene samples
prepared with dithioesters 2-7 and with benzyl dithio-
benzoate, 23, as a comparison. As it can be seen, Mn
increases linearly with time in all cases, as is expected
for properly controlled polymerization processes. It
should however be noted that while the increase of Mn
in the case of the three pyridinyl derivatives 2-4 is
comparable to that observed with dithiobenzoate 23, it
becomes much slower for the pyridinyl N-oxides 5 and
6 and even more so for 7. An examination of Figure 4
also reveals that the polydispersity index, although
initially rather good for all the heteroaromatic CTAs,
increases as the polymerization proceeds whereas a
decrease of Mw/Mn is expected for a well-behaved RAFT
agent, similarly to that observed with 23.
On a similar basis, we identify as radical 20 the
species responsible for the final spectrum observed in
the electrochemical reduction of compounds 2 and 5. The
observation of 20 upon reduction of 5 and of 21 upon
reduction of 7 indicates that deoxygenation must also
take place at some stage.
The observation of such radical dianions in the
reduction of dithioesters is not unprecedented. Indeed,
the radical dianion (EtO)2P(O)C(S)S•2- was previously
A comparison between 23 and dithioesters 4 and 7
makes more evident the different behavior of the
heteroaromatic compounds as RAFT agents. Figure 5a
shows the SEC curves observed for polystyrene samples

7616 Alberti et al.
Macromolecules, Vol. 38, No. 18, 2005
Figure 5. SEC (size exclusion chromatography) curves (a) and
variation of the ratio Mn/yield with time (b) for polystyrene
samples obtained at 80 °C after 30 h using compounds 4
(central trace and 2), 7 (leftmost trace and 4), and 23
(rightmost trace and b) as RAFT agents.
Figure 4. Plot of the number-average molar mass Mn (a) and
polydispersity index (b) as a function of time for the AIBN-
initiated polymerization of styrene with RAFT agents 2 ([),
3 (9), 4 (2), 5 (]), 6 (0), 7 (4), and 23 (b) at 60 °C.
obtained when using the three RAFT CTAs. It can be
seen that in the three cases the curves are fairly narrow
corresponding to a polydispersity index of 1.20, 1.25, and
1.15 for 4, 7, and 23, respectively, and that there is a
decrease of Mn in the opposite order.
The different behavior of dithioesters 4, 7, and 23 is
also evident when the ratio between Mn and the yield
is plotted as a function of time (Figure 5b). In the case
of the N-oxide 7 the ratio that represents the maximum
attainable molecular mass decreases rather steeply. In
the case of the pyridinyl dithioester 4 the value of the
ratio is smaller and decreases more slowly, whereas for
compound 23 the value is fairly constant and close to
100 000, a value that corresponds to a degree of poly-
merization of 1000, which is similar to the molar ratio
between styrene and the RAFT agents used in the
polymerizations.
Even more revealing is the trend of the polymeriza-
tion yield with time for the three CTAs that is shown
in Figure 6. It can be seen that the polymerization
process slows down significantly when replacing 23 with
4 and virtually comes to a halt when compound 7 is
used.
It is well established that a critical factor determining
the rate of RAFT polymerization is the rate of fragmen-
tation of the “dormant” radical species (1 in Scheme 1)
and hence its stability. As already mentioned, the capto-
dative effect plays a very important role in the stabiliza-
tion of this species. From the reduction potentials
reported in Table 1 it is evident that the extent of capto-
dative stabilization should be greater for pyridinyl
dithioesters than for dithiobenzoates, and this may
result in a lower fragmentation rate of 1 and in a slower
Figure 6. Plot of yield vs time for the AIBN-initiated
polymerization of styrene at 60 °C using 4 (2), 7 (4), and 23
(b) as RAFT transfer agents.
overall polymerization process. On the other hand, the
difference between the reduction potentials of the py-
ridinyl dithioesters (e.g., 4) and those of the correspond-
ing N-oxides (e.g., 7) seems too small to account for the
dramatic reduction of the polymerization observed when
using the latter compounds as RAFT agent. Additional
effects other than the simple difference in capto-dative
stabilization must then be taken into account.
Focusing the attention on the “dormant” radicals 1,
the remarkable difference between those derived from
pyridinyl dithioesters and those from the corresponding
N-oxides (indicated in Scheme 4 as 24 and 25, respec-
tively) cannot go unnoticed.

Macromolecules, Vol. 38, No. 18, 2005
Using Pyridinyl Dithioesters as RAFT CT Agents 7617
Scheme 4
As already stressed above and at variance with
radical 24, radical 25 can be viewed as a diunsaturated
nitroxide, which makes the contribution of mesomeric
structure B more relevant than in the case of 24. This
is also supported by DFT calculations on the structur-
ally related radical 10. Because of the well-known ability
of nitroxides to couple with alkyl radicals, radical 25
might then act as a trap for the growing polystyryl
radical to give alkoxyamine 26, thereby hampering the
polymerization process. True enough, under appropriate
conditions the formation of 26 may be a reversible
process, and indeed the reversible coupling of relatively
stable nitroxides with growing polymer radicals is also
largely used to control radical polymerization, the
process being known as NMP (nitroxide-mediated po-
lymerization).38,39 Fragmentation of the oxygen-carbon
bond in 26 requires more drastic conditions, i.e. higher
temperatures, than necessary for the cleavage of the
sulfur-carbon bond in 25 or, more generally in the
dormant radicals 1; actually, when polymerization of
styrene in the presence of 7 was carried out at temper-
atures T > 110 °C, a higher yield was obtained but the
polydispersity index was very poor, i.e. greater than 2.
percent of conversion. A tentative explanation of the
inefficiency of these last compounds involves termina-
tion through coupling between the propagating radical
and an in situ generated nitroxide. Thus, the failure of
dithioisonicotinate N-oxide 7 in acting as an efficient
RAFT mediator should not be attributed to excessive
stabilization of the intermediate radical 1 and could not
be foretold on the basis of its reduction potential.
References and Notes
(1) Le, T. P.; Moad, G.; Rizzardo, E.; Thang, S. H. PCT Int. Appl.
WO 9801478 A1 980115, Jan 25, 1998.
(2) Chiefari, J.; Chong, Y. K.; Ercole, F.; Krsitina, J.; Jeffery, J.;
Le, T. P. T.; Mayadunne, R. T. A.; Meijs, G.; Moad, C. L.;
Moad, G.; Rizzardo, E.; Thang, S. H. Macromolecules 1998,
31, 5559-5562.
(3) Quinn, J. F.; Rizzardo, E.; Davis, T. P. Chem. Commun. 2001,
1044-1045.
(4) Goto, A.; Sato, K.; Tsujii, Y.; Fukuda, T.; Moad, G.; Rizzardo,
E.; Thang, S. H. Macromolecules 2001, 34, 402-408.
(5) 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.
(6) Chong, Y. K.; Krstina, J.; Le, T. P. T.; Moad, G.; Postma, A.;
Rizzardo, E.; Thang, S. H. Macromolecules 2003, 36, 2256-
2272.
(7) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Chong, J. K.;
Moad, G.; Thang, S. H. Macromolecules 1999, 32, 6977-6980.
(8) Mayadunne, R. T. A.; Rizzardo, E.; Chiefari, J.; Krstina, J.;
Moad, G.; Postma, A.; Thang, S. H. Macromolecules 2000,
33, 243-245.
Indirect support of this explanation may be provided
by the higher yields observed when replacing 7 with
N-oxides 5 and 6. Indeed, the nitroxidic structure is not
possible for the “dormant” adduct of the meta-dithioester,
and although possible for that of the ortho-derivative
5, its formation might be prevented by steric factors.
(9) Vana, P.; Davis, T. P.; Barner-Kowollik, C. Macromol. Theory
Simul. 2002, 11, 823-835.
(10) Kwak, Y.; Goto, A.; Tsujii, Y.; Murata, Y.; Komatsu, K.;
Fukuda, T. Macromolecules 2002, 35, 3026-3029.
(11) Coote, M. L.; Radom, L. J. Am. Chem. Soc. 2003, 125, 1490-
1491.
(12) Coote, M. L. Macromolecules 2004, 37, 5023-5031.
(13) Viehe, H. G.; Mere´nyi, R.; Stella, L.; Janousek, Z. Angew.
Chem., Int. Ed. Engl. 1979, 18, 7-932.
Concluding Remarks
The greater electron-withdrawing ability of the py-
ridinyl and pyridinyl N-oxide residue with respect to
that of phenyl results in lower reduction potentials for
the heteroaromatic dithioesters than for dithioben-
zoates. The radical anions of these compounds as well
as their spin adducts resulting via thiophilic addition
of free radicals were readily observed by ESR thanks
to the stabilizing effect exerted by the pyridinyl and
pyridinyl N-oxide residues that is similar to, or slightly
greater than, that of phosphoryl and thiophosphoryl
groups.
Pyridinyl dithioesters control the radical polymeri-
zation of styrene, although with a poorer control of the
molar mass distribution as compared to benzyl dithioben-
zoate and with an evident retardation. This may be
related to the rate of bond dissociation and re-formation
with respect to the propagation rate. Pyridinyl N-oxide
dithioesters, on the other hand, are not suitable as
RAFT CTAs as they stop the polymerization at a few
(14) Viehe, H. G.; Janousek, Z.; Mere´nyi, R. Acc. Chem. Res. 1985,
18, 148-154.
(15) Coote, M. L.; Henry, D. J. Macromolecules 2005, 38, 1415-
1433.
(16) Laus, M.; Papa, R.; Sparnacci, K.; Alberti, A.; Benaglia, M.;
Macciantelli, D. Macromolecules 2001, 34, 7269-7275.
(17) Alberti, A.; Benaglia, M.; Laus, M.; Macciantelli, D.; Spar-
nacci, K. Macromolecules 2003, 36, 736-740.
(18) Abrunhosa, I.; Gulea, M.; Masson, S. Synthesis 2004, 928-
934.
(19) Trost, B.; Braslau, R. J. Org. Chem. 1988, 53, 532-537.
(20) Grierson, D. S.; Bettiol, J.-L.; Buck, I.; Husson, H.-P.;
Rubiralta, M.; Diez, A. J. Org. Chem. 1992, 57, 6414-6421.
(21) Crispino, G.; Breslow, R. J. Org. Chem. 1992, 57, 1849-1855.
(22) Jarvis, B. B.; Marien, B. A. J. Org. Chem. 1977, 42, 2676-
2680.
(23) Garreau, D.; Save´ant, J.-M.; Stella, L. Electroanal. Chem.
1972, 35, 309-331.

7618 Alberti et al.
Macromolecules, Vol. 38, No. 18, 2005
(24) Lucarini, M.; Luppi, B.; Pedulli, G. F.; Roberts, B. P. Chem.s
Eur. J. 1999, 5, 2048-2054.
(31) Guerra, M. J. Phys. Chem. A 1999, 103, 5983-5988.
(32) Mayer, R.; Scheihauer, S.; Kunz, D. Chem. Ber. 1966, 99,
1393-1413.
(25) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(26) Lee, C.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785-
789.
(33) Alberti, A.; Guerra, M.; Hapiot, P.; Lequeux, T.; Macciantelli,
D.; Masson, S. Phys. Chem. Chem. Phys. 2005, 7, 250-257.
(27) 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.;
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.; Chen, W.; Wong, M. W.; Andres, J.
L.; Gonzalez, C.; Head-Gordon, M.; Replogle, E. S.; Pople, J.
A. GAUSSIAN 98, Revision A.7; Gaussian, Inc.: Pittsburgh,
PA, 1998.
(34) Voss, J.; von Bulov, C.; Drews, T.; Mischke, P. Acta Chem.
Scand., Ser. B 1983, 37, 519-525.
(35) Alberti, A.; Benaglia, M.; Hapiot, P.; Hudson, A.; Le Cous-
tumer, G.; Macciantelli, D.; Masson, S. J. Chem. Soc., Perkin
Trans. 2 2000, 1908-1913.
(36) Iwai, K.; Takemura, F.; Furue, M.; Nozakura, S.-I. Bull.
Chem. Soc. Jpn. 1984, 57, 763-767.
(37) Forrester, A. R. In Landolt-Bo¨rnstein, New Series, Group II;
Fischer, H. Ed.; Springer-Verlag: Berlin, 1989; Vol. 17, Part
d2.
(38) Moad, G.; Anderson, A. G.; Krstina, J.; Moad, C. L.; Rizzardo,
E.; Spurlinig, T. H.; Thang, S. H. In Controlled Radical
Polymerzation; ACS Symposium Series No. 685; Matyjasze-
wski, K., Ed.; American Chemical Society: Washington, DC,
1997; pp 332-360.
(39) Advances in Controlled/Living Radical Polymerisation; ACS
Symposium Series No. 854; Matyjaszewski, K., Ed.; American
Chemical Society: Washington, DC, 2003.
(28) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28,
213-222.
(29) Francl, M. M.; Petro, W. J.; Hehre, W. J.; Binkley, J. S.;
Gordon, M. S.; De Frees, D. J.; Pople, J. A. J. Chem. Phys.
1982, 77, 3654-3665.
(30) Clark, T.; Chandrasekhar, J.; Schleyer, P. v. R. J. Comput.
Chem. 1983, 4, 294-301.
MA050652D