Hydroxy- and Silyloxy-Substituted TEMPO Derivatives for the Living Free-Radical Polymerization of Styrene and n-Butyl Acrylate: Synthesis, Kinetics, and Mechanistic Studies

Article abstract of DOI:10.1021/ja037948o

The synthesis of new 2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO) styryl derivatives as mediators for the living free-radical polymerization is described. Two of the α-methyl groups at the 2- and 6-position of the parent TEMPO styryl alkoxyamine have been

Full text of DOI:10.1021/ja037948o

Published on Web 12/05/2003
Hydroxy- and Silyloxy-Substituted TEMPO Derivatives for the
Living Free-Radical Polymerization of Styrene and n-Butyl
Acrylate: Synthesis, Kinetics, and Mechanistic Studies
Christoph Alexander Knoop and Armido Studer*
Contribution from the Department of Chemistry, Philipps-UniVersity Marburg,
35032 Marburg, Germany
Received August 15, 2003; E-mail: studer@mailer.uni-marburg.de
Abstract: The synthesis of new 2,2,6,6-tetramethylpiperidin-1-oxyl (TEMPO) styryl derivatives as mediators
for the living free-radical polymerization is described. Two of the R-methyl groups at the 2- and 6-position
of the parent TEMPO styryl alkoxyamine have been replaced by hydroxymethyl and silyloxymethyl groups.
To further increase the steric hindrance around the alkoxyamine oxygen atom, the remaining two methyl
groups have been substituted with larger ethyl groups. Styrene polymerizations using hydroxy-substituted
TEMPO derivatives are fast, but are not well-controlled. As previously shown for other OH-substituted
alkoxyamines, intramolecular H-bonding leads to an acceleration of the C-O bond homolysis and, hence,
to an acceleration of the polymerization process. However, the OH groups also increase the alkoxyamine
decomposition rate constant. The kinetics of the C-O bond homolysis have been determined using EPR
spectroscopy. Decomposition studies have been conducted with the aid of 1H NMR spectroscopy. In contrast
to the OH-substituted alkoxyamines, highly hindered silyloxy-substituted TEMPO alkoxyamines turned out
to be excellent mediator/initiators for the controlled styrene polymerization. Polystyrene with Mn of up to
80 000 g/mol and narrow polydispersities (PDI) has been prepared using the new alkoxyamines. Reactions
have been conducted at 105 °C; however, even at 90 °C controlled but slow polymerizations can be
achieved. Furthermore, and more importantly, poly(n-butyl acrylates) with narrow PDIs (<1.15) have been
prepared at 105 °C with the new alkoxyamines. Controlled acrylate polymerization can be conducted at
temperatures as low as 90 °C. The silylated alkoxyamines presented belong to the most efficient initiator/
mediators for the controlled acrylate polymerization known to date. The effect of the addition of free nitroxide
on the acrylate polymerization is discussed. Moreover, the synthesis of diblock copolymers with narrow
PDIs is described.
radical, respectively, is of great importance in these processes.4,5
We6 and others7,8 have shown that H-bonding in nitroxides leads
to a stabilization of the nitroxide, and this in turn alters the
equilibrium constant between the alkoxyamine and the corre-
Introduction
Nitroxide-mediated polymerizations (NMP),1 atom transfer
radical polymerizations (ATRP),2 and RAFT polymerizations3
provide polymers with polydispersities (PDI) below the theoreti-
cal limit for a conventional radical polymerization (1.5). The
NMP process is controlled by the persistent radical effect (PRE)4
and is based on the reversible formation of a dormant
alkoxyamine from the corresponding nitroxide and the chain-
growing polymer radical. The equilibrium in this controlled
radical polymerization lies far on the side of the dormant
alkoxyamine, ensuring a low concentration of free radicals
during the polymerization. The equilibrium constant between
the nitroxide-capped polymer and the free nitroxide and polymer
(5) (a) Goto, A.; Terauchi, T.; Fukuda, T.; Miyamoto, T. Macromol. Rapid
Commun. 1997, 18, 673-681. (b) Fukuda, T.; Goto, A. Macromol. Rapid
Commun. 1997, 18, 683-688. (c) Skene, W. G.; Belt, S. T.; Conolly, T.
J.; Hahn, P.; Scaiano, J. C. Macromolecules 1998, 31, 9103-9105. (d)
Ohno, K.; Tsujii, Y.; Miyamoto, T.; Fukuda, T.; Goto, M.; Kobayashi, K.;
Akaike, T. Macromolecules 1998, 31, 1064-1069. (e) Bon, S. A. F.;
Chambard, G.; German, A. L. Macromolecules 1999, 32, 8269-8276. (f)
Goto, A.; Fukuda, T. Macromol. Chem. Phys. 2000, 201, 2138-2142. (g)
Fischer, H.; Souaille, M. Chimia 2001, 55, 109-113. (h) Marque, S.;
Fischer, H.; LeMercier, C.; Tordo, P. Macromolecules 2000, 33, 4403-
4410. (i) Drockenmuller, E.; Catala, J.-M. Macromolecules 2002, 35, 2461-
2466. (j) Bertin, D.; Chauvin, F.; Marque, S.; Tordo, P. Macromolecules
2002, 35, 3790-3791. (k) LeMercier, C.; Acerbis, S.; Bertin, D.; Chauvin,
F.; Gigmes, D.; Guerret, O.; Lansalot, M.; Marque, S.; Le Moigne, F.;
Fischer, H.; Tordo, P. Macromol. Symp. 2002, 182, 225-247. (l) Cun-
ningham, M. F.; Tortosa, K.; Lin, M.; Keoshkerian, B.; Georges, M. K. J.
Polym. Sci., Part A: Polym. Chem. 2002, 40, 2828-2841. Work on other
nitroxide structures: (m) Cresidio, S. P.; Aldabbagh, F.; Busfield, W. K.;
Jenkins, I. D.; Thang, S. H.; Zayas-Holdsworth, C.; Zetterlund, P. B. J.
Polym. Sci., Part A: Polym. Chem. 2001, 39, 1232-1241. (n) Blomberg,
S.; Ostberg, S.; Harth, E.; Bosman, A. W.; Van Horn, B.; Hawker, C. J. J.
Polym. Sci., Part A: Polym. Chem. 2002, 40, 1309-1320. (o) Dervan, P.;
Aldabbagh, F.; Zetterlund, P. B.; Yamada, B. J. Polym. Sci., Part A: Polym.
Chem. 2003, 41, 327-334.
(1) (a) Solomon, D. H.; Rizzardo, E.; Cacioli, P. U.S. Patent 4,581,429, 1985;
P. Eur. Pat. Appl. 135280, 1985 (Chem. Abstr. 1985, 102, 221335q). (b)
Georges, M. K.; Veregin, R. P. N.; Kazmaier, P. M.; Hamer, G. K.
Macromolecules 1993, 26, 2987-2988. For a review, see: (c) Hawker, C.
J.; Bosman, A. W.; Harth, E. Chem. ReV. 2001, 101, 3661-3688.
(2) (a) Matyjaszewski, K.; Xia, J. Chem. ReV. 2001, 101, 2921-2990. (b)
Kamigaito, M.; Ando, T.; Sawamoto, M. Chem. ReV. 2001, 101, 3689-
3746.
(3) Barner-Kowollik, C.; Davis, T. P.; Heuts, J. P. A.; Stenzel, M. H.; Vana,
P.; Whittaker, M. J. Polym. Sci., Part A: Polym. Chem. 2003, 41, 365-
375.
(4) Fischer, H. Chem. ReV. 2001, 101, 3581-3610.
(6) Marque, S.; Fischer, H.; Baier, E.; Studer, A. J. Org. Chem. 2001, 66,
1146-1156.
9
10.1021/ja037948o CCC: $25.00 © 2003 American Chemical Society
J. AM. CHEM. SOC. 2003, 125, 16327-16333
16327

A R T I C L E S
Knoop and Studer
Scheme 2. Preparation of the Various Alkoxyamines Studied
Scheme 1. Preparation of Hindered Piperidines via a Dianion
Double-Alkylation Strategy
sponding nitroxide and C-centered radical. Nitroxides capable
of forming intramolecular H-bonds have been shown to be very
efficient mediators for the controlled/living radical polymeri-
zation of n-butyl acrylate.8
Herein we present results on the synthesis of new hydroxy-
and silyloxy-substituted 2,2,6,6-tetramethylpiperidin-1-oxyl
(TEMPO) derivatives. Their efficiency as mediators for the
polymerization of styrene and n-butyl acrylate will be discussed.
Furthermore, rate constants for the C-O bond homolysis of
the new alkoxyamines derived from these nitroxides and
decomposition studies of the new alkoxyamines will be pre-
sented.
Results and Discussion
Preparation of the Alkoxyamines. Esterification of diacid
1, reduction, and subsequent Boc protection afforded meso-
diester 2 (60%, Scheme 1).9 Double alkylation of 2 provided
trans-3a (14%) and cis-3b (65%).10 Ethylated compounds 4a,b
were prepared in analogy.
N-Deprotection (HBr/AcOH) followed by oxidation
(Na2WO4, H2O2) afforded the nitroxides 5a,b10 and 6a,b in
excellent yields (Scheme 2). Alkoxyamine synthesis according
to a literature procedure11 provided 7a, 7b, 8a, and 8b in mod-
erate to good yields. The alkoxyamines 7a and 8a were obtained
as a 1:1 mixture of inseparable diastereoisomers. LiAlH4 reduc-
tion gave diols 9a,b and 10a,b. The diols were protected using
standard procedures (f 11-14). The known1c alkoxyamines
16 and 17 were included in this study for comparison. The
relative configuration of all new alkoxyamines could unambigu-
ously be assigned on the basis of X-ray analysis of 9b and 10b.12
Polymerization of Styrene. The initial polymerizations were
conducted in sealed tubes using 1 mol % of the alkoxyamine
initiator at 125 °C and were stopped after 6 h. The conversion
was determined gravimetrically. If conversion exceeded 95%
after 6 h, polymerizations were repeated and were stopped after
1.5 to 3 h. The PDI and the molecular weight of the polymers
were analyzed using size exclusion chromatography (SEC). The
results are summarized in Table 1.
Polymerizations using the diesters 7a,b were not controlled
(PDI > 1.5, entries 1 and 2). The alkoxyamines 8a,b provided
higher conversions; however, control was still not satisfactory
(entries 3 and 4). With trans-diol initiator 9a, a conversion of
74% was achieved in just 1.5 h; however, the measured
polydispersity was rather broad (1.25, entry 5). The reaction
with cis-diol 9b was slower, but a narrow PDI was obtained
(1.14, entry 6). The ethyl-substituted trans-diol 10a provided a
fast, but not well-controlled polymerization (entry 7).13 The best
result was obtained with the cis-ethyl-substituted diol 10b, where
reaction occurred fast and controlled (entry 11). Unfortunately,
(7) (a) Matyjaszewski, K.; Gaynor, S.; Greszta, D.; Mardare, D.; Shigemoto,
T. J. Phys. Org. Chem. 1995, 8, 306-315. (b) Matyjaszewski, K.; Gaynor,
S.; Greszta, D.; Mardare, D.; Shigemoto, T. Macromol. Symp. 1995, 98,
73-89. (c) Goto, A.; Kwak, Y.; Yoshikawa, C.; Tsujii, Y.; Sugiura, Y.;
Fukuda, T. Macromolecules 2002, 35, 3520-3525.
(8) Harth, E.; Van Horn, B.; Hawker, C. J. Chem. Commun. 2001, 823-824.
(9) Goldspink, N. J.; Simpkins, N. S.; Beckmann, M. Synlett 1999, 1292-
1294.
(10) Einhorn, J.; Einhorn, C.; Ratajczak, F.; Pierre, J.-L. J. Chem. Soc., Chem.
Commun. 1995, 1029-1030.
(11) Matyjaszewski, K.; Woodworth, B. M.; Zhang, X.; Gaynor, S. G.; Metzner,
Z. Macromolecules 1998, 31, 5955-5957.
(12) Crystallographic data (excluding structure factors) for the structures reported
in this paper have been deposited with the Cambridge Crystallographic
Data Center as private communication no. CCDC-217340 (9b) and CCDC-
217341 (10b) (Klaus Harms, 2003). Copies of the data can be obtained
free of charge on application to CCDC, 112 Union Road, Cambridge
CB21EZ, U.K. (fax: (+44) 1233 336-033; e-mail: deposit@ccdc.cam.ac.uk).
(13) Marque, S.; Sobek, J.; Fischer, H.; Kramer, A.; Nesvadba, P.; Wunderlich,
W. Macromolecules 2003, 36, 3440-3442.
9
16328 J. AM. CHEM. SOC. VOL. 125, NO. 52, 2003

Hydroxy- and Silyloxy-Substituted TEMPO Derivatives
A R T I C L E S
Table 1. Polymerization of Styrene Using Alkoxyamines 7-14, 16, and 17 as Mediator/Initiators under Different Conditions
alkoxyamine
(mol %)
time
(h)
temp
(°C)
M
(g/mol)
M
(g/mol)
conversion
(%)
alkoxyamine
(mol %)
time
(h)
temp
(°C)
M
(g/mol)
M
n,th
conversion
(%)
n,exp
n,th
n,exp
entry
PDI
entry
(g/mol)
PDI
1
2
3
4
5
6
7
8
7a (1%)
7b (1%)
8a (1%)
8b (1%)
9a (1%)
6
6
6
6
1.5
3
2
2
2
2
3
3
3
3
6
6
125
125
125
125
125
125
125
125
125
125
125
125
125
125
125
125
4600
7100
6300
6100
10300
6500
10100
15400
37000
46000
5900
14100
27100
57000
4800
4500
6000
6400
7400
7700
1.64
1.57
1.32
1.22
1.25
1.14
1.29
1.24
1.24
1.36
1.13
1.16
1.22
1.32
1.19
1.19
43
58
61
71
74
67
95
70
62
43
68
63
50
46
50
69
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
13a (1%)
13a (0.4%)
13a (0.2%)
13a (0.1%)
13a (1%)
13a (0.4%)
13a (0.1%)
13a (0.4%)
13b (1%)
13b (0.1%)
13a,b (0.1%)
14 (1%)
2
3
3
125
125
125
125
105
105
105
90
125
105
105
125
105
125
125
105
7300
18200
36000
48200
9500
20300
78200
17800
9200
67400
69900
8100
22100
1800
6600
16400
30200
47900
8900
18700
70500
13800
6800
69800
68700
6500
19000
2000
1.12
1.13
1.17
1.20
1.12
1.11
1.20
1.10
1.12
1.22
1.21
1.11
1.17
1.30
1.16
1.17
63
63
58
46
85
72
68
53
65
67
66
62
73
19
63
51
3
15
20
21
48
2
21
21
2
9b (1%)
7000
9900
10a (1%)
10a (0.4%)
10a (0.2%)
10a (0.1%)
10b (1%)
10b (0.4%)
10b (0.2%)
10b (0.1%)
11 (1%)
18200
21100
44700
7100
16400
26100
48200
5200
9
10
11
12
13
14
15
16
14 (0.4%)
16 (1%)
17 (1%)
20
6
6
5000
6400
6600
5300
12 (1%)
7100
7200
17 (1%)
15
it turned out that 10a and 10b are not perfectly suited for the
controlled preparation of high molecular weight polystyrene.
For both alkoxyamines, an increased PDI (1.3 to 1.4) was
obtained for polymers with Mn values above 30 000 g/mol
(entries 8-10, 12-14). We speculated that the OH groups lead
to decreased nitroxide, alkoxyamine stability (stability studies
will be discussed below), or both. Therefore, we studied
reactions with O-protected derivatives of alkoxyamines 9b, 10a,
and 10b (f 11, 12, 13a,b). Polymerization of styrene with bis-
(methyl ether) 11 occurred slower as compared to 9b, bearing
two free hydroxyl groups (compare entry 11 with 15). Further-
more, PDI slightly increased. The TBDMS-prototected alkoxy-
amine 12 is more efficient than 11 (entry 16), and the ethyl-
congener 13a turned out to be a highly efficient initiator/
mediator (63%, 2 h, PDI ) 1.12, entry 17). A similar result
was obtained for the cis-derivative 13b (entry 25).
To document the efficiency of 13a,b, polymerizations were
repeated with established alkoxyamines for comparison. Using
styryl TEMPO 16, we obtained a conversion of only 19% with
a PDI of 1.30 under the same conditions (Table 1, entry 30).
With the well-known Hawker-Braslau alkoxyamine 17,14a
a
slower polymerization and a slightly higher PDI were achieved
(entries 31 and 32).
13a can be used for the preparation of high molecular weight
polystyrenes at 125 °C (Table 1, entries 18-20). At 105 °C,
even better results were obtained (Mn up to 78 200 g/mol with
PDI below 1.20, entries 21-23). The diastereoisomer 13b is
also a very efficient initiator/mediator (entry 26). Polymerization
using a mixture of the trans-13a and cis-13b (64:36 mixture)
provided the same result (entry 27). It is important to note that
it is not necessary to separate the cis- and trans-compounds
during synthesis. Therefore, the most demanding task in large-
scale alkoxyamine synthesis, namely the chromatographic
separation of 4a and 4b, is not necessary! Bis(benzyl ether) 14
works as well as TBDMS ether 13a, showing that the Si group
has no special influence (entries 28 and 29). Thus, the
“expensive” TBDMS groups can be replaced by benzyl groups
without losing initiator/mediator efficiency. Furthermore, we
found that controlled styrene polymerization with 13a can be
conducted at temperatures as low as 90 °C (entry 24, 53%, PDI
) 1.10).
Figure 1. (a) Molecular weight and PDI vs monomer conversion. (b)
Monomer consumption vs time (styrene, 105 °C, 0.4 mol % 13a).
tion as a function of time, and we analyzed the molecular weight
as a function of monomer conversion (Figure 1). The linear
increase of ln([M]0/[M]) vs time and linear increase of molecular
weight vs conversion as well as the decrease of PDI at increasing
conversion prove the controlled character of the polymeri-
zation.
Polymerization of n-Butyl Acrylate. We then studied the
n-butyl acrylate polymerization using 13a,b and 10a. To date,
NMP of acrylates is still a challenge.1c Polymerizations were
performed in neat n-butyl acrylate as described in the Supporting
To prove the controlled/living character of the 13a-mediated
styrene polymerization, we determined the monomer consump-
9
J. AM. CHEM. SOC. VOL. 125, NO. 52, 2003 16329

A R T I C L E S
Knoop and Studer
Table 2. Polymerization of n-Butyl Acrylate Using Alkoxyamines 10a, 13a,b, and 17 as Mediator/Initiators under Different Conditions
alkoxyamine
(mol %)
nitroxide 18a
(mol %)
time
(h)
temp
(°C)
M
(g/mol)
M
(g/mol)
conversion
(%)
n,exp
n,th
entry
PDI
1
2
3
4
5
6
7
8
10a (0.2%)
10a (0.2%)
13a (0.4%)
13a (1%)
13a (1%)
13a (1%)
13a (1%)
13a (1%)
13a (1%)
13a (1%)
13a (1%)
13a (0.4%)
13a (0.4%)
13a (0.1%)
13a (1%)
13a (1%)
13b (1%)
13a,b (1%)
17 (1%)
6
16
15
24
24
24
24
24
24
24
24
24
42
42
64
96
24
24
6
125
105
125
105
105
105
105
105
105
105
105
105
105
105
90
43900
71700
37100
13300
6000
57000
47500
28900
9200
4100
5500
6700
8100
8100
8300
1.79
1.27
1.55
1.20
1.12
1.13
1.14
1.15
1.16
1.17
1.19
1.23
1.13
1.22
1.29
1.16
1.16
1.15
1.21
90
74
90
72
32
43
52
63
63
65
66
67
69
45
57
39
65
67
48
0.05%
0.04%
0.02%
0.01%
0.008%
0.006%
0.004%
8200
10100
11500
11800
12300
12600
31600
29100
64700
25000
17300
11000
11800
9500
9
10
11
12
13
14
15
16
17
18
19
8500
21500
22100
57700
18300
12600
8300
0.02%
0.005%
0.01%
0.01%
0.01%
0.05%a
90
105
105
125
8500
6200
a
The nitroxide derived from alkoxyamine 17 was used. For the preparation of the nitroxide, see ref 14a.
Scheme 3. Synthesis of Nitroxide 18a via Thermolysis of
Alkoxyamine 13a
mediators for the controlled acrylate polymerization known to
date.1c,14,15 To the best of our knowledge, poly(n-butyl acrylate)
with polydispersities around 1.1 prepared via NMP have so far
only been reported for noncyclic alkoxyamines.1c,14,15 To
underline the efficiency of the new initiators, we repeated the
acrylate polymerization under the same conditions using one
of the best alkoxyamines (f 17) known to date. At 125 °C 17
provided worse results under our conditions (entry 19). More-
over, polymerizations at 105 °C were not reproducible using
17. Controlled NMP of acrylates is not possible using TEMPO
initiator 16.
Information. Diol 10a was not able to control the polymerization
at 105 and 125 °C (Table 2, entries 1 and 2). At 125 °C the
13a-mediated acrylate polymerization was also not controlled
(entry 3). However, at 105 °C we obtained poly(n-butyl acrylate)
with a narrow PDI (entries 4 and 12). The dormant polymeric
alkoxyamine derived from 13a is obviously not sufficiently
stable at 125 °C; however, at 105 °C activity and stability for
efficiently controlled acrylate polymerization is high enough.
All the reported successfully controlled nitroxide-mediated
acrylate polymerizations have been conducted in the presence
of 5 mol % free nitroxide (with respect to the alkoxyamine).14
Therefore, we repeated the polymerization in the presence of
nitroxide 18a which was readily prepared via thermolysis of
13a in the presence of oxygen (Scheme 3). Polymerization was
slower, but PDI further decreased (Table 2, entry 5). We were
also able to prepare n-butyl acrylate with a Mn of 64 700 g/mol
in a controlled manner (entries 13 and 14). As for the styrene
polymerizations, the diastereoisomer 13b shows similar activity
than 13a (entry 17). A mixture of 13a,b (64:36 mixture)
provided the same result (entry 18). Controlled polymerization
can even be conducted at 90 °C (entries 15 and 16). It is
important to note that 13a,b belong to the most efficient initiator/
The effect of additional nitroxide on the conversion and on
the PDI has been theoretically analyzed by Fukuda14d and
Fischer.14f At large additional nitroxide concentrations, the
conversion can be described using eq 1 ([Y]0 ) concentration
of added free nitroxide; K ) equilibrium constant between active
and dormant chains (K ) kd/kc); kc ) rate constant for the cross-
coupling of the C-centered radical with the nitroxide; kd )
dissociation rate constant for the C-O bond homolysis; kp )
propagation rate constant; [I]0 ) initial alkoxyamine concentra-
tion). At lower nitroxide concentrations, however, the conversion
has to be described using eq 2, as suggested by Fukuda (kt )
termination rate constant).14d
ln([M]0/[M]) ) kpK([I]0/[Y]0)t
(1)
ln([M]0/[M]) ) (kp/2ktK[I]0){(3ktK2[I]0 t +
2
[Y]03)2/3 - [Y]0 } (2)
2
According to eq 2, to increase the conversion the amount of
added nitroxide has to be decreased. Indeed, in our experiments
a lowering of the free nitroxide concentration from 0.04 mol %
to 0.004 mol % resulted in an increase of the conversion (from
43 to 65%, Table 2, entries 6-11). As expected from the
theoretical work, PDI slightly decreased along with the increase
of the nitroxide concentration (from 1.13 to 1.19). However,
even at the lowest nitroxide concentration tested, polymerization
(14) (a) Benoit, D.; Chaplinski, V.; Braslau, R.; Hawker, C. J. J. Am. Chem.
Soc. 1999, 121, 3904-3920. (b) Benoit, D.; Hawker, C. J.; Huang, E. E.;
Lin, Z.; Russell, T. P. Macromolecules 2000, 33, 1505-1507. (c) Benoit,
D.; Grimaldi, S.; Robin, S.; Finet, J.-P.; Tordo, P.; Gnanou, Y. J. Am. Chem.
Soc. 2000, 122, 5929-5939. (d) Fukuda, T.; Goto, A.; Ohno, K. Macromol.
Rapid Commun. 2000, 21, 151-165. (e) Lacroix-Desmazes, P.; Lutz, J.-
F.; Chauvin, F.; Severac, R.; Boutevin, B. Macromolecules 2001, 34, 8866-
8871. (f) Souaille, M.; Fischer, H. Macromolecules 2002, 35, 248-261.
(g) Tang, C.; Kowalewski, T.; Matyjaszewski, K. Macromolecules 2003,
36, 1465-1473. (h) Schierholz, K.; Givehchi, M.; Fabre, P.; Nallet, F.;
Papon, E.; Guerret, O.; Gnanou, Y. Macromolecules 2003, 36, 5995-5999.
(15) Sterically highly hindered 4-oxo-TEMPO-derived alkoxyamines for the
controlled/living radical polymerization of n-butyl acrylate providing
polyacrylates with PDIs below 1.15: Wetter, C.; Gierlich, J.; Knoop, C.;
Mu¨ller, C.; Schulte, T.; Studer, A. Chem.-Eur. J., in press.
9
16330 J. AM. CHEM. SOC. VOL. 125, NO. 52, 2003

Hydroxy- and Silyloxy-Substituted TEMPO Derivatives
A R T I C L E S
the concentration of the released nitroxide was measured by
EPR spectroscopy, as previously described.5e,h,6,17 The experi-
mental cleavage rate constant kd was calculated using eq 3
(conversions up to 30%). The activation energy Ea was estimated
from the rate constants using A ) 2.4 × 1014 s-1 6
.
[nitroxide]∞ - [nitroxide]t
ln
) -kdt
(3)
(
)
[nitroxide]∞
As expected from the polymerization results described above,
the smallest kd values were obtained for the bis(esters) 7 and 8
(Table 3, entries 1-4). In agreement with the H-bonding
concept,6-8 larger rate constants were measured for alkoxyamines
bearing free hydroxyl groups (9, 10, entries 5-8). For the methyl
and the ethyl series, the cis/trans-isomers showed similar rate
constants. As expected for steric reasons, the ethyl-substituted
alkoxyamines homolyze faster than the corresponding methyl
compounds (compare entries 5 and 6 with 7 and 8). The
importance of steric effects on the rate of the C-O bond
cleavage has previously been discussed.5h,6,13,15 In agreement
with the polymerizations, large rate constants were measured
for the silyloxy alkoxyamines 13a and 13b (entries 11 and 12).
Probably, steric effects overcompensate the importance of
H-bonding upon switching from 10a,b to 13a,b. Furthermore,
there is no special Si effect: the bis(benzyl ether) 14 homolyzes
as fast as the corresponding Si-substituted alkoxyamines 13
(compare entries 11 and 12 with 13). As comparison, for the
Hawker-Braslau alkoxyamine 17 a higher Ea was measured for
the homolysis under the same conditions (127.1 kJ/mol).5h
From the EPR experiments we could obtain important
information on the stability of the various nitroxides (Figure
3). The EPR data of the new nitroxides were directly obtained
from the kinetic experiments and are summarized in Table 3.
For an alkoxyamine leading to a stable nitroxide, 90-95% of
nitroxide was liberated during the EPR experiment. However,
in EPR experiments on alkoxyamines leading to unstable
nitroxides the concentration of the released nitroxide did not
exceed 50% of the initial alkoxyamine concentration. As
example, results of EPR experiments on the homolysis of the
hydroxy-substituted alkoxyamine 10b and the corresponding
silyl-protected congener 13b are presented in Figure 4. The
concentration of liberated nitroxide 18b derived from bis(silyl
ether) 13b reached the maximum after 175 s (about 93%
nitroxide liberated). No decomposition of 18b was noticed
during the following 1250 s. However, a different picture was
observed for the hydroxy derivative 10b. The maximum
concentration of 20b (35%) was reached after 120 s.18 After
120 s, the nitroxide concentration steadily decreased and then
Figure 2. ln([M]0/[M]) as a function of time for the determination of K
according to eq 1 (bulk polymerization at 105 °C using 1 mol % 13a and
0.5 mol % of 18a, kp ) 7.1 × 104 M-1 s-1).14e
was still well-controlled. The best result in terms of conversion
and narrow PDI was obtained using 1 mol % nitroxide 18a with
respect to alkoxyamine 13a.
Lacroix-Desmazes used eq 1 to determine the equilibrium
rate constant K for the DEPN system (DEPN ) N-tert-butyl-
N-(1-diethylphosphono-2,2-dimethylpropyl) nitroxide).14e The
ln([M]0/[M]) was plotted as a function of time at high concen-
tration of added nitroxide. K was directly obtained from the
slope (1.6 × 10-11 M at 105 °C). In analogy, we determined K
for the n-butyl acrylate polymerization mediated by 18a at 105
°C (Figure 2). A value of 4.2 × 10-12 M was extracted. This is
in the same range as the K value obtained for the DEPN system,
which is among the best noncyclic mediators known to date
for the acrylate polymerization.14e From the K value and the
experimentally determined kd for the homolysis of 15 (3.1 ×
10-4 s-1 at 105 °C, vide infra), kc was readily calculated (kc )
1.4 × 108 M-1 s-1).
Block Copolymerizations. We then investigated the forma-
tion of diblock copolymers using alkoxyamine 13a. To this end,
polystyrene or poly(n-butyl acrylate) prepared with alkoxyamine
13a was used as a macroinitiator for the controlled styrene and
acrylate polymerization. Poly(n-butyl acrylate)-block-(polysty-
rene) with a Mn of 45 000 g/mol and a narrow PDI (1.16) was
obtained at 105 °C using a poly(n-butyl acrylate) macroinitiator
(0.2 mol %, Mn ) 10 800 g/mol, PDI ) 1.13, 15 h, 54%).
(Polystyrene)-block-poly(n-butyl acrylate) could be prepared
using a polystyrene macroinitiator (0.2 mol %, Mn ) 9500
g/mol, PDI ) 1.12) at 105 °C for 24 h; however, conversion
remained low (18%). The block copolymer was formed with a
narrow PDI (PDI ) 1.18, Mn ) 23 000 g/mol). Problems in
the formation of (polystyrene)-block-poly(n-butyl acrylate) using
polystyrene macroinitiators have previously been reported for
noncyclic alkoxyamines.14a In addition, we used a polystyrene
macroinitiator (0.2 mol %, Mn ) 9500 g/mol, PDI ) 1.12) for
the neat styrene polymerization (15 h, 105 °C, 56% conversion).
Polystyrene with a narrow PDI (1.17) and increased Mn was
obtained (Mn ) 38 300 g/mol), documenting the living character
of 13a-mediated styrene polymerizations.
(16) Sobek, J.; Martschke, R.; Fischer, H. J. Am. Chem. Soc. 2001, 123, 2849-
2857.
(17) (a) Kovtun, G. A.; Aleksandrov, A. L.; Golubev, V. A. Bull. Acad. Chim.
USSR, DiV. Chem. 1974, 10, 2115-2121. (b) Howard, J. A.; Tait, J. C. J.
Org. Chem. 1978, 43, 4279-4283. (c) Grattan, D. W.; Carlsson, D. J.;
Howard, J. A.; Wiles, D. M. Can. J. Chem. 1979, 57, 2834-2842.
(18) For the most efficient alkoxyamines, the nitroxide is liberated within the
first 250 s. Our experimental setup allows us to reach the target temperature
of the EPR sample in about 90 s, as independently determined. Therefore,
for the systems fast homolyzing our kd, Ea may be at the lower limit. To
address this problem, we determined Ea for alkoxyamine 13a at 378 and
363 K where temperature adjustment is not an issue. In fact, similar Ea’s
were calculated using 2.4 × 1014 s-1 as the Arrhenius factor (Ea (407 K)
) 122.2 kJ/mol, Ea (378 K) ) 122.4 kJ/mol, Ea (363 K) ) 121.9 kJ/mol),
indicating that the error is small using our experimental setup, even for
the most efficient alkoxyamines. We thank a reviewer for drawing our
attention to this problem.
Kinetics of the C-O Bond Homolysis: EPR Experiments.
The kinetic experiments were conducted in tert-butylbenzene
at 407 K. Oxygen was used to scavenge the styryl radical, and
9
J. AM. CHEM. SOC. VOL. 125, NO. 52, 2003 16331

A R T I C L E S
Knoop and Studer
Table 3. Results of the Kinetic Experiments and Decomposition Studies and EPR Data (g ) 2.006, kd and kdec,nitroxide Were Determined at
407 K; kdec,alkoxyamine Was Determined at 398 K)
entry
alkoxyamine
kd (s-1
)
E (kJ/mol)a
k
dec, alkoxyamine (s-1
)
nitroxide
kdec,nitroxide (M-1 s-1 b
)
aN (G)
a
1
2
3
4
5
6
7
8
7a
7b
8a
8b
9a
2.7 × 10-4
6.6 × 10-4
6.4 × 10-4
9.9 × 10-4
1.8 × 10-3
2.3 × 10-3
6.1 × 10-3
5.5 × 10-3
1.3 × 10-3
2.1 × 10-3
3.5 × 10-2
2.1 × 10-2
1.9 × 10-2
3.9 × 10-3
138.5
135.5
135.6
134.1
132.1
131.3
128.0
128.4
133.3
131.6
122.2
123.8
124.2
129.5
3.6 × 10-6
3.5 × 10-6
7.1 × 10-6
1.5 × 10-5
4.0 × 10-5
1.7 × 10-5
8.2 × 10-5
8.9 × 10-5
4.7 × 10-6
3.8 × 10-6
1.9 × 10-5
1.5 × 10-5
1.2 × 10-5
1.9 × 10-5
5a
5b
6a
stable
stable
stable
stable
25
60
110
135
stable
stable
stable
stable
stable
stable
14.01
14.01
13.44
13.58
14.87
14.66
15.23
15.30
15.30
15.16
14.37
14.16
14.30
14.37
6b
19a
19b
20a
20b
21
9b
10a
10b
11
9
10
11
12
13
14
12
22
13a
13b
14
18a
18b
23
15
18a
a
b
Ea was calculated using A ) 2.4 × 1014 s-1
.
Stable indicates no significant decomposition of the nitroxide during the EPR experiment.
Scheme 4. Suggested Mechanism for the Decomposition of the
HO-Functionalized Nitroxides.
Figure 3. Several nitroxides studied.
of these considerations, the evolution of the nitroxide concentra-
tion during the EPR experiment can be described by the kinetic
eq 4 ([Y] ) nitroxide concentration; [I] ) alkoxyamine
concentration; kd ) rate constant for the C-O bond homolysis;
kdec,nitroxide ) rate constant for the bimolecular nitroxide decom-
position).
d[Y]/dt ) kd[I] - 2kdec,nitroxide[Y]2
(4)
With the experimentally determined kd, we simulated the
evolution of the nitroxide concentration using Powersim, a
program for modeling nonlinear dynamics.20 The nitroxide
concentration evolution ([20b]) for the homolysis experiment
of 10b could be perfectly modeled using the experimentally
determined rate constant of 5.5 × 10-3 s-1 for the initial C-O
bond homolysis and 135 M-1 s-1 as a rate constant for the
bimolecular nitroxide decomposition (Figure 4, Table 3, entry
8).21 In analogy, the rate constants for the decomposition of
nitroxides 19a,b and 20a were estimated. For 19a and 20a, a
nearly perfect fit of the experimental data was obtained with
the simulation, whereas for 19b the experimental data could
not be ideally described using eq 4 (see the Supporting
Information). To our knowledge, this is the first report where
nitroxide decomposition rate constants and alkoxyamine C-O
bond homolysis rate constants have been extracted from a single
experiment.
Figure 4. Evolution of the experimentally determined nitroxide concentra-
tion during homolysis of alkoxyamines 10b and 13b at 407 K. The line
presents the concentration profile simulated using 5.5 × 10-3 s-1 as rate
constant for the initial C-O bond homolysis and 135 M-1 s-1 as the rate
constant for the bimolecular decomposition.
leveled off after 450 s. Similar nitroxide concentration profiles
were obtained for the HO-substituted alkoxyamines 9a,b and
10a, indicating decomposition of the nitroxide during the EPR
experiment (see the Supporting Information). We believe that
the rate-determining step of the nitroxide decomposition is a
bimolecular process, as suggested in Scheme 4.19 On the basis
(19) The decomposition of 4-oxo-TEMPO also occurs via a bimolecular
process: (a) Murayama, T.; Yosshika, T. Bull. Chem. Soc. Jpn. 1969, 42,
2307-2309. (b) Rozantsev, E. G.; Sholle, V. D. Synthesis 1971, 401-
414. (c) Han, C. H.; Drache, M.; Baethge, H.; Schmidt-Naake, G.
Macromol. Chem. Phys. 1999, 200, 1779-1783.
(20) Powersim Software Home Page. http://www.powersim.com (Nov 2003).
(21) The kinetic experiment with alkoxyamine 10b was repeated using different
concentrations (1 × 10-4, 1.25 × 10-4, 1.5 × 10-4, and 2 × 10-4 M).
Similar nitroxide decomposition rate constants were extracted from these
experiments.
9
16332 J. AM. CHEM. SOC. VOL. 125, NO. 52, 2003

Hydroxy- and Silyloxy-Substituted TEMPO Derivatives
A R T I C L E S
From these EPR experiments we suggest that the OH-bearing
nitroxides decompose via a self-reaction (bimolecular process)
and are therefore not sufficiently stable for NMP. Two questions
remain to be answered: Why do OH groups accelerate the self-
reaction and why are the O-protected alkoxyamines stable? We
believe that intermolecular H-bonding between two nitroxides
leads to a preorganization which in turn accelerates the
decomposition. Indeed, in the X-ray structure of the cis-
alkoxyamines 9b and 10b intermolecular H-bonding can be
identified.12
Thermal Stability of the Alkoxyamines. The thermal
decomposition of a polymeric alkoxyamine leading to a
terminally unsaturated polymer and the corresponding hydroxy-
lamine is an important side reaction in NMP. This process can
occur via a â-hydrogen atom transfer from the transient polymer
radical to the nitroxide or via a nonradical direct ionic
elimination.7c,22 We studied the thermal decomposition of the
various alkoxyamines to provide styrene and the corresponding
hydroxylamine. The alkoxyamine was dissolved in a NMR tube
in perdeutero p-xylene (0.03 to 0.06 M). The degassed sample
was heated to 398 K within the cavity of a 500 MHz 1H NMR
spectrometer, and the decomposition was followed by monitor-
ing the decrease of the alkoxyamine signals as well as the
increase of the styrene resonances. The signal of the benzylic
H atom at around 4.7 was used to estimate the alkoxyamine
concentration. Similar experiments have previously been
described.7c,22,23 Spectra were recorded every 5 min. The
decomposition rate constants (kdec,alkoxyamine) for the various
alkoxyamines were determined using eq 5 ([S] ) styrene
concentration; [A] ) alkoxyamine concentration) and are
summarized in Table 3.7c
than the trans-alkoxyamine 9a. The ethyl-substituted alkoxyamines
decompose 2-5 times faster than the methyl compounds. Thus,
the introduction of hydroxy functions and sterically demanding
ethyl groups leads to the desired larger C-O bond homolysis
rate constants; however, the stability of the alkoxyamines
simultaneously decreases. The moderate stability of the
alkoxyamine 10b explains why polymerizations leading to
polymers with molecular weights above 40 000 are not well-
controlled. For an ideal alkoxyamine initiator/mediator ho-
molysis should be fast, and at the same time decomposition must
be slow. This is difficult to achieve because assuming that part
of the decomposition occurs via disproportionation, the decom-
position rate constant must correlate with the homolysis rate
constant. However, for our Si-substituted alkoxyamines 13a,b,
homolysis is about 3 times faster and decomposition is about 5
times slower than for the corresponding hydroxy derivatives
10a,b. This explains why 13a,b are so efficient for NMP. The
bis(benzyl ether) 14 decomposes with a similar rate constant
as the TBDMS-protected alkoxyamines (entry 13), again show-
ing that the Si-substituent has no special effect on the kinetics
of the corresponding alkoxyamine.
Conclusions
We developed a straightforward synthesis for the preparation
of a series of new bulky TEMPO styryl derivatives bearing
R-hydroxymethyl, R-silyloxymethyl, or ethyl groups instead of
methyl groups at positions 2 and 6. The silyloxy-substituted
TEMPO derivatives have been shown to be highly efficient
mediators for the controlled/living polymerization of styrene
and n-butyl acrylate. Diblock copolymers with narrow PDIs have
successfully been prepared with the silylated alkoxyamines. We
have shown that controlled polymerization can be conducted
at 90 °C. The new alkoxyamines belong to the most efficient
initiator/mediators for the controlled acrylate polymerization
ln([S]/[A] + 1) ) kdec,alkoxyamine
t
(5)
1
known to date.1c,14,15 Kinetic EPR experiments and H NMR
Smallest kdec,alkoxyamine values were measured for the ester-
functionalized alkoxyamines (Table 3, entries 1-4). As expected
from the EPR experiments, significantly faster decompositions
were obtained for the hydroxy-modified TEMPO derivatives
9a,b and 10a,b. In the ethyl series, both isomers decomposed
at similar rates; however, for the methyl-substituted alkoxyamines
9a,b the trans-isomer is about 2 times less stable than the cis-
isomer (entries 5-8). This agrees with the styrene polymeri-
zation results where the cis-isomer 9b provided better results
alkoxyamine decomposition studies have been applied to explain
the polymerization results.
Acknowledgment. We thank the Deutsche Forschungsge-
meinschaft (STU 280/1-1) for funding. Dr. Olaf Burghaus and
Tobias Schulte are acknowledged for conducting the EPR
experiments. Dr. Klaus Harms is acknowledged for determining
the X-ray structures. Furthermore, we thank the reviewers for
helpful suggestions.
Supporting Information Available: Experimental procedures
and analytical data of all new compounds (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
(22) (a) Ananchenko, G. S.; Fischer, H. J. Polym. Sci., Part A: Polym. Chem.
2001, 39, 3604-3621. (b) Schulte, T.; Studer, A. Macromolecules 2003,
36, 3078-3084.
(23) The thermal stability experiments are not really a mimic of NMP conditions
since no monomer is present. Decomposition during polymerization is
therefore far slower.
JA037948O
9
J. AM. CHEM. SOC. VOL. 125, NO. 52, 2003 16333