Steric and Electronic Effects in Cyclic Alkoxyamines - Synthesis and Applications as Regulators for Controlled/Living Radical Polymerization

Article abstract of DOI:10.1002/chem.200305427

The synthesis of new six- and seven-membered cyclic alkoxyamines bearing ethyl groups at the α-N position of the alkoxyamines is described. The key step in the synthesis of the sterically hindered six-membered cyclic alkoxyamines is a Wadsworth-Horner-Emm

Full text of DOI:10.1002/chem.200305427

FULL PAPER
Steric and Electronic Effects in Cyclic Alkoxyamines–Synthesis and
Applications as Regulators for Controlled/Living Radical Polymerization
Christian Wetter, Johannes Gierlich, Christoph Alexander Knoop, Christoph M¸ller,
Tobias Schulte, and Armido Studer*[a]
Abstract: The synthesis of new six- and
seven-membered cyclic alkoxyamines
bearing ethyl groups at the a-N posi-
tion of the alkoxyamines is described.
The key step in the synthesis of the
sterically hindered six-membered cyclic
alkoxyamines is a Wadsworth Horner
Emmons olefination with bisphospho-
nate 1. The seven-membered cyclic alk-
oxyamines were prepared from the
corresponding six-membered keto alk-
oxyamines by ring-enlargement with
trimethylsilyl(TMS)-diazomethane. The
use of the new alkoxyamines as regula-
tors/initiators for radical polymeriza-
tion is discussed. Efficient controlled
and living polymerization of styrene
and n-butyl acrylate was obtained with
the six-membered tetraethyl alkoxy-
amine 13. Controlled polymerizations
can be conducted even at 908C. In ad-
dition, alkoxyamine 13 can be used for
the preparation of AB diblock and
ABA triblock copolymers with narrow
polydispersities. The influence of the
replacement of methyl groups in the a-
position of the N atom in cyclic alkoxy-
amines by larger ethyl groups on the
styrene polymerization (reaction time,
ꢀ
PDI, kinetics of the C O bond homol-
ysis) is discussed. In addition, thermal
decomposition of the new alkoxy-
amines was studied. Furthermore, the
synthesis of N,N-bissilylated alkoxy-
amines is described. The silylated alk-
oxyamines are not suitable as regula-
tors/initiators for the controlled/living
ꢀ
radical polymerization. The C O
bonds in silylated alkoxyamines are
ꢀ
stronger than the C O bonds in analo-
Keywords: alkoxyamines
¥
block
gous N,N-dialkylated alkoxyamines.
The experimental results are verified
by calculations with Gaussian98 (A.9).
copolymers ¥ kinetics ¥ nitroxides ¥
radical polymerization
Introduction
The equilibrium constant between the nitroxide-capped pol-
ymer and the free nitroxide and polymer radical, respective-
ly, is of great importance. Various parameters, such as H
bonding,[5] steric and electronic effects among others, influ-
ence the equilibrium.[6,7] Based on calculations, Moad and
During the last ten years, radical polymerization has gained
renewed interest. Different methods have been introduced
that allow controlled living radical polymerizations. Nowa-
days, polymers with narrow polydispersities (PDI<1.2) can
be prepared by nitroxide-mediated polymerizations
(NMPs),[1] atom-transfer radical polymerizations (ATRPs),[2]
or reversible addition fragmentation chain transfer (RAFT)
polymerizations.[3] The NMP and ATRP processes are con-
trolled by the persistent radical effect (PRE).[4] For NMP,
reversible formation of a dormant alkoxyamine from the
corresponding nitroxide and the chain-growing polymer rad-
ical is the key to the success of the controlled/living poly-
merization. The equilibrium in the NMP lies well towards
the side of the dormant alkoxyamine, thus ensuring a low
concentration of free radicals during the polymerization.
ꢀ
Rizzardo predicted small C O bond dissociation energies
for silylated alkoxyamines of type A (Scheme 1).[8] More-
Scheme 1. Variation of the steric and electronic effects of the N substitu-
ents in alkoxyamines.
[a] Dipl. Chem. C. Wetter, Dipl. Chem. J. Gierlich,
Dipl. Chem. C. A. Knoop, Dipl. Chem. C. M¸ller,
Dipl. Chem. T. Schulte, Prof. Dr. A. Studer
Fachbereich Chemie der Universit‰t Marburg
Hans-Meerwein-Strasse, 35032Marburg (Germany)
Fax : (+49)6421-282-5629
over, they showed that increasing the steric bulk around the
ꢀ
N atom in the alkoxyamines should decrease the C O bond
strength (!B). Very recently, Fischer discussed steric effects
of ring substituents in cyclic alkoxyamines on the equilibri-
um constant based on experimental results.[9] In agreement
1156
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
DOI: 10.1002/chem.200305427
Chem. Eur. J. 2004, 10, 1156 1166

1156 1166
with the theoretical predictions, an increase in the size of
LiBr/1,8-diazabicyclo[5.4.0]undec-7-ene
(DBU)/THF,[11]
ꢀ
the substituents leads to a decrease of the C O bond disso-
ciation energy.
KHCO3/tBuOH/H2O, Ba(OH)2/THF/H2O, and LDA/THF
the desired olefin 2 was not observed. With NaH in THF
the product was obtained in moderate yields; however, up-
scaling was not possible. It turned out that dianion forma-
tion is essential for this particular WHE reaction. Thus,
treatment of 1 with LDA and subsequent addition of BuLi
afforded the corresponding dianion that was reacted with 3-
pentanone to give phosphonate 2, which was used for the
second olefination without any purification (Scheme 2). Un-
fortunately, one-pot bis-olefination could not be accomplish-
ed. The second WHE reaction with 3-pentanone was diffi-
cult to achieve. Olefination with LiBr/DBU/THF,[11] LDA/
THF, NaH/THF, MgBr2/NEt3/THF,[12] and Ba(OH)2/THF/
Herein we present results on the synthesis of new sterical-
ly hindered styryl-TEMPO derivatives (TEMPO=2,2,6,6-
tetramethylpiperidinyl-1-oxyl). In addition, the synthesis of
silylated alkoxyamines is described. The efficiency of the
new alkoxyamines as regulators for the polymerization of
styrene and n-butyl acrylate is discussed. In addition, the
preparation of AA and AB diblock as well as of ABA tri-
block copolymers is presented. Furthermore, rate constants
ꢀ
for the C O bond homolysis of the new alkoxyamines de-
rived from these nitroxides and decomposition studies of
the new alkoxyamines are given. Finally, calculations on the
ꢀ
C O bond dissociation energy of silylated alkoxyamines is
described.
Results and Discussion
Preparation of the alkoxyamines: Bisphosphonate 1 was
prepared according to a literature procedure in two steps
starting from commercially available methallyl dichloride.[10]
Wadsworth Horner Emmons (WHE) olefination with 3-
pentanone was attempted under different conditions. With
Abstract in German: In dieser Arbeit wird die Herstellung
von zyklischen 6- und 7-Ring-Alkoxyaminen beschrieben, die
in a-Stellung zum Stickstoffatom Ethylgruppen tragen. Der
Schl¸sselschritt in der Synthese dieser sterisch gehinderten Alk-
oxyamine beinhaltet eine Wadsworth Horner Emmons-
Olefinierung am Bisphosphonat 1. Die 7-gliedrigen Homolo-
gen wurden ausgehend von den entsprechenden Piperidino-
nen hergestellt. Die Ringerweiterung wurde mit TMS-Diazo-
methan durchgef¸hrt. Es wird ¸ber die Verwendung der
neuen Alkoxyamine als Regulatoren/Initiatoren in der radika-
lischen Polymerisation berichtet. Effiziente kontrollierte und
lebende Polymerisation von Styrol und n-Butylacrylat wurde
mit dem Tetraethylpiperidinon-Alkoxyamin 13 erzielt. Kont-
rollierte Polymerisationen konnten sogar bei 908C durchge-
f¸hrt werden. Au˚erdem kann das Alkoxyamin 13 f¸r den
Aufbau von AB-Diblock- und ABA-Triblock-Copolymeren
mit niedrigen Polydispersit‰ten verwendet werden. Es wird
diskutiert, welchen Einfluss der Austausch von Methyl-durch
Ethylgruppen in a-Stellung zum Stickstoff in zyklischen Alk-
oxyaminen auf die Styrolpolymerisation aus¸bt (Reaktions-
ꢀ
dauer, PDI, Kinetiken der C O-Bindungshomolyse). Die
thermische Zersetzung der neuen Alkoxyamine wurde unter-
sucht. Des Weiteren wird die Herstellung von N,N-bissilyliert-
en Alkoxyaminen beschrieben. Die silylierten Alkoxyamine
sind als Initiatoren/Regulatoren f¸r kontrollierte/lebende Pol-
ꢀ
ymerisationen nicht geeignet. Die C O-Bindungen in sily-
ꢀ
lierten Alkoxyaminen sind st‰rker als die C O-Bindungen in
entsprechenden dialkylierten Alkoxyaminen. Die experimen-
tellen Ergebnisse wurden mit Hilfe von Berechnungen mit
Gaussian98 (A.9) best‰tigt.
Scheme 2. Synthesis of the nitroxides 6a,b, 8, 10, and 12.
1157
Chem. Eur. J. 2004, 10, 1156 1166
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

A. Studer et al.
FULL PAPER
H2O failed. Pleasingly, a smooth olefination was observed
with CsOH in a benzene/H2O mixture. The WHE products
were isolated in 31% overall yield as a mixture of the de-
sired ketone 3a along with the isomerized b,g-unsaturated
ketones 4 (cis/trans mixture of isomers; ratio 3a/4 = 1:6).
Prolonged heating under the reaction conditions did not
alter the isomer ratio (thermodynamic product control).
Separation of the isomers was not necessary because isomer-
ization occurred under the conditions applied for the subse-
quent cyclization reaction (concentrated NH4OH, 1058C).
Heating of the 3a,4 mixture in concentrated NH4OH at
1058C in a sealed tube for 20 h afforded piperidinone 5
(40%, 59% based on recovered 3a and 4), that was oxi-
dized with Na2WO4/H2O2 in MeOH/H2O[13] to give nitroxide
6a in 87% isolated yield. Nitroxide 6b was prepared in
analogy from 2 via ketone 3b.
Hydroxy derivative 8 was obtained via reduction of 5
(LAH, !7) and subsequent oxidation (95% overall). Ring-
enlargement of 5 with TMSCHN2/BF3¥OEt2 (!9)[7] and oxi-
dation afforded keto nitroxide 10 in 33% overall yield.
LAH reduction of 9 gave alcohol 11 that was directly oxi-
dized to give nitroxide 12 (96%). The alkoxyamines 13 17
were prepared from 1-phenylethyl bromide and the corre-
sponding nitroxides according to the Matyjaszewski protocol
(Scheme 3).[14] For a proper discussion of the steric effects of
the ring substituents of our new alkoxyamines on the poly-
merizations, results previously obtained with the smaller
methyl-substituted alkoxyamines 18 22 are also considered
in the present study.
Scheme 3. New sterically hindered alkoxyamines 13 17 and the corre-
sponding known methyl derivatives 18 22.
The silylated alkoxyamines were prepared from commer-
cially available 1,2-dibromobenzene. Formation of the bis-
Grignard reagent from 1,2-dibromobenzene and silylation
with diethylchlorosilane or diisopropylchlorosilane afforded
the corresponding dialkylarylsilanes (Scheme 4). The chloro-
silanes were readily obtained from the silanes upon treat-
ment with Cl2 (!23,24).[15,16a] Silylation of 1-phenylethoxya-
mine[16b] with 23 and 24 to give 25 and 26 was best achieved
under microwave irradiation in DMF with Et3N and catalyt-
ic amounts of DMAP.
Scheme 4. Synthesis of the silylated alkoxyamines 25 and 26.
more, excellent control of the polymerization was achieved
(PDI
= 1.12, Table 1, entry 1). As expected, the ethyl
groups accelerate the polymerization process: styrene poly-
merization with the corresponding smaller methyl-substitut-
Polymerization of styrene with alkoxyamines 13 17: The
polymerizations were conduct-
ed in sealed tubes with 1% of
Table 1. Polymerization of styrene using alkoxyamines 13 17.
the alkoxyamine initiator at
90 1258C and were stopped
after 6 56 h. The conversion
was determined gravimetrical-
ly. The polydispersity index
(PDI) and the molecular
weight of the polymers were
analyzed by means of size-ex-
Entry
Alkoxyamine
(mol%)
Temp
[8C]
Time
[h]
Mn,exp
Mn,theory
PDI
Conversion
[%]
Ref.
[gmolꢀ1
]
[gmolꢀ1
]
[a]
[a]
[a]
[a]
[a]
[b]
[b]
[b]
[b]
[a]
[a]
[a]
[a]
[a]
[a]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
13 (1)
14 (1)
15 (1)
16 (1)
17 (1)
18 (1)
19 (1)
20 (1)
21 (1)
13 (0.2)
13 (1)
13 (0.2)
13 (0.1)
13 (1)
22 (1)
125
125
125
125
125
125
125
125
125
125
105
105
105
90
6
6
6
6
6
6
6
6
6
10900
11700
7400
7700
10600
3000
2200
9000
3300
43500
9400
8200
7100
6800
6900
5800
2900
2700
8700
3100
38500
7200
38000
73900
7600
2000
1.12
1.36
1.14
1.13
1.64
1.23
1.27
1.10
1.23
1.32
1.09
1.16
1.24
1.08
1.30
79
68
65
66
56
28
26
84
30
74
69
73
71
73
19
clusion
chromatography
(SEC). The results are sum-
marized in Table 1.
6
24
24
24
56
6
The fastest polymerization
was observed for the six-mem-
bered keto alkoxyamine 13 for
which a 79% conversion was
obtained after 6 h. Further-
37500
62400
10500
1800
125
[a] This work. [b] Ref. [7]
1158
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 1156 1166

Steric and Electronic Effects in Cyclic Alkoxyamines
1156 1166
ed six-membered keto alkoxyamine regulator 18 afforded
only 28% conversion under the same conditions (Table 1,
entry 6). The well-known styryl-TEMPO alkoxyamine 22
gave a conversion of only 19% with a PDI of 1.30 under the
same conditions (Table 1, entry 15). The same trend was ob-
served for the six-membered hydroxy-alkoxyamines 14 and
19 . With the larger ethyl-substituted initiator, 14, polymeri-
zation was faster than with 19 (Table 1, entries 2, 7). Howev-
er, the control of the polymerization was not satisfactory for
the hydroxy-alkoxyamines 14 and 19 (PDI>1.2). We have
previously[7] shown that hydroxy-substituted cyclic alkoxy-
amines are not sufficiently stable under the reaction condi-
tions. This is probably the reason for the modest control ob-
tained with regulator 14. Stability studies are discussed
below. The methyl-ethyl-substituted initiator 15 provided a
fast polymerization and a good control (Table 1, entry 3). A
good result was obtained with the seven-membered alkoxy-
amine 16 (66% conversion, PDI = 1.13, Table 1, entry 4).
Contrary to our expectations, a higher conversion was meas-
ured for the smaller methyl congener 20 (Table 1, entry 8).
As for the six-membered alkoxyamines, polymerization was
not controlled with the hydroxy derivative 17 (Table 1,
entry 5). From these initial results we can conclude that, for
six-membered alkoxyamines, the substitution of methyl
groups with larger ethyl groups leads to an improvement of
the initiator/regulator efficiency. However, for the seven-
membered systems, the size of the substituents did not influ-
ence the polymerization process to a large extent. In fact,
the tetramethyl-substituted seven-membered alkoxyamine
20 turned out to be a slightly better initiator/regulator than
the ethyl congener 16 for the styrene polymerization. Fur-
thermore, keto groups in the rings are tolerated, whereas 4-
hydroxy-substituted cyclic alkoxyamines provide moderate
results.
Figure 1. a) Monomer conversion versus time (styrene, 1058C, 1 mol%
13). b) Molecular weight vs monomer conversion (styrene, 1058C,
1 mol% 13, 15 h).
versus time and molecular weight versus monomer con-
sumption proves the controlled character of the polymeriza-
tion.
It is well-known that styryl-TEMPO 22 is not able to con-
trol the acrylate polymerization.[1c] To date, controlled nitro-
xide-mediated polymerization of acrylates is only possible
with two types of nitroxides.[6g,17,18] To control the acrylate
polymerization, however, sacrificial nitroxide has to be
added to the reaction mixture in these systems. Therefore,
we focused on the polymerization of n-butyl acrylate with
alkoxyamine 13 and without additional nitroxide. The poly-
merizations were conducted at 105 or 1258C in neat n-butyl
acrylate (Table 2). We were very pleased to observe that
polymerization was efficient and controlled at 1258C
(Table 2, entry 1). Even better results were obtained on low-
ering the reaction temperature to 1058C. Reaction for 32h
provided poly(n-butyl acrylate) with a PDI of 1.12with Mn
of 18600 gmolꢀ1 (Table 2, entry 2). It is important to note
that no nitroxide has to be added to the reaction mixture.
To our knowledge, this is the first report of a successfully
controlled polymerization of an acrylate mediated by a
cyclic nitroxide.[19]
Further polymerization studies were conducted with the
efficient alkoxyamine 13. We first attempted to prepare high
molecular-weight polystyrene using 13. The reaction was
performed with 0.2mol% of 13 under otherwise identical
conditions (Table 1, entry 10). Disappointingly, the polymer-
ization was not well controlled (PDI
= 1.32, Mn =
43500 gmolꢀ1). We repeated the polymerization at 1058C
(Table 1, entry 11, 1 mol% 13). To our delight, polymeriza-
tion over a period of 24 h afforded polystyrene with a PDI
of 1.09 (69% conversion). Therefore, lowering of the tem-
perature is important for a successful polymerization. High
molecular weight polystyrene (up to 62500 gmolꢀ1) with a
narrow PDI can be prepared at 1058C (Table 1, entries 12,
13). Even at 908C, controlled (but slow) polymerization of
styrene can be performed with our new alkoxyamine 13
(Table 1, entry 14).
To prove the controlled/
living character of the 13-
mediated styrene polymeriza-
tion, we determined the con-
version as a function of time
and we analyzed the molecular
weight as a function of mono-
mer conversion (Figure 1). The
linear increase of ln([Mo]/[M])
Table 2. Polymerization of n-butyl acrylate with 13 under different conditions.
Entry
Alkoxyamine
(mol%)
Temp
[8C]
Time
[h]
Mn,exp
Mn,theory
PDI
1.18
Conversion
[%]
[gmolꢀ1
]
[gmolꢀ1
]
1
2
3
4
5
13 (1)
13 (1)
13 (0.2)
13 (0.1)
13 (1)
125
105
105
105
90
9
20500
11400
89
84
70
3218600
32
3291300
96
10600
73000
83300
12300
18.132
53800
1.34
9000
1.21
65
1.13
1159
Chem. Eur. J. 2004, 10, 1156 1166
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

A. Studer et al.
FULL PAPER
All the following experiments were therefore conducted
at 1058C. With 0.2mol% 13, high conversion (84%) and
good control (PDI = 1.21) was obtained after 32 h (Table 2,
entry 3). High molecular-weight poly(n-butyl acrylate) was
prepared with 0.1 mol% of the alkoxyamine (Mn
=
91300 gmolꢀ1, Table 2, entry 4). However, the polydispersity
slightly increased. We also showed that the butyl acrylate pol-
ymerization can be conducted at 908C (Table 2, entry 5, 96 h,
PDI = 1.13, 70% conversion).
We next studied the preparation of block copolymers with
alkoxyamine 13. To this end, alkoxyamine-terminated poly-
styrene and poly(n-butyl acrylate) were used as macroinitia-
tors. Polymerizations were conducted under bulk conditions
at 105 and 1258C. The results are summarized in Table 3.
Styrene polymerization at 1258C with a polystyrene macroi-
nitiator (PDI = 1.10, Mn = 9300 gmolꢀ1) afforded polystyr-
ene with an increased PDI (PDI = 1.45, Table 3, entry 1).
A slightly better result was obtained if a poly(n-butyl acry-
late) was used as a macroinitiator (Table 3, entry 2). The
poly(n-butyl acrylate)-block-polystyrene (P(n-BA)-b-PS) di-
block copolymer was formed with a PDI of 1.28. We
thought that decreasing the reaction temperature would fur-
ther improve the results. Indeed, styrene polymerization
Figure 2. GPC traces of the PS-b-P(n-BA) diblock copolymer (trace b)
and the P(n-BA)-b-PS-b-(n-BA) triblock copolymer (trace d) and the
corresponding macroinitiators (trace a: PS macroinitiator; trace c: P(n-
BA)-b-PS-macroinitiator).
(Table 3, entry 6, Figure 2, trace d). In analogy, we prepared
a PS-b-P(n-BA)-b-PS triblock copolymer with excellent con-
trol (PDI = 1.14, Table 3,entry 7). We can conclude that
AB diblock and ABA triblock copolymers with narrow
PDIs can be efficiently prepared with alkoxyamine 13.
Polymerization studies with
the silylated alkoxyamines: We
Table 3. Preparation of di- and triblock copolymers using alkoxyamine 13.
Entry
Macroinitiator
(Mn , PDI, mol%)
Monomer
T
Time Mn,exp
Mn,theory PDI
Conversion
[%]
then studied styrene polymeri-
zations with the silylated al-
koxyamines 25 and 26. The
polymerizations were conduct-
ed under the conditions descri-
bed above. Disappointingly,
neither 25 nor 26 were able to
control the polymerization
[8C] [h]
[gmolꢀ1
]
[gmolꢀ1
]
1
2P(
3
PS (9300, 1.10, 0.2%)
n-BA) (18600, 1.12, 0.2%)
PS (5000, 1.12, 1%)
styrene
styrene
styrene
125
125
105 12
6
6
33600
42000
19400
48900
70700
10300
1.45
58
1.28[a] 50
1.08
1.15
51
57
[b]
4
PS (10500, 1.08, 0.2%)
n-butyl acry- 105 3266400
74600
1.08
late
[a]
[b]
5
6
P(n-BA) (17800, 1.13, 0.2%)
P(n-BA)-b-PS (16700, 1.12,
0.5%)
styrene
105 12
39200
57400
1.14
42
31
n-butyl acry- 105 3230900 47020
late
7
PS-b-P(n-BA) (6600, 1.15,
0.5%)
styrene
105 16
25400
22200
1.14[a] 61
(PDI>2.0).
Polymerizations
are governed by the autopoly-
merization of styrene.[20] The
alkoxyamines are only specta-
tors in these polymerizations!
Contrary to the calculations
ꢀ
[a] Measured with polystyrene standard solution. [b] Measured with poly(methylmethacrylate) standard solu-
tion.
with a PS-macroinitiator at 1058C afforded polystyrene with
a narrow PDI (PDI = 1.08, Table 3, entry 3). Furthermore,
a polystyrene-block-poly(n-butyl acrylate) (PS-b-P(n-BA))
diblock copolymer with an Mn of 66400 gmolꢀ1 and a
narrow PDI (1.08) was prepared at 1058C starting with a PS
macroinitiator (Table 2, entry 4). GPC traces of the PS mac-
roinitiator and the PS-b-P(n-BA) diblock copolymer are
presented in Figure 2(traces a and b). Nitroxide-mediated
polystyrene-block-poly(n-butyl acrylate) diblock formation
was not successful with the alkoxyamines known to date.[17]
P(n-BA)-b-PS diblock copolymer formation worked very
well with alkoxyamine 13 (Table 3, entry 5). Encouraged by
these results, we then focused on the formation of ABA tri-
block copolymers. To this end, a P(n-BA)-b-PS macroinitia-
tor (Mn = 39200 gmolꢀ1, PDI = 1.15) was used for the
neat acrylate polymerization at 1058C. We were very
pleased to observe that ABA triblock copolymerization
reported in the literature,[8] C O bonds in silylated alkoxya-
mines are obviously stronger than in the corresponding N,N-
dialkylated alkoxyamines. Therefore, we decided to reinves-
ꢀ
tigate the C O bond strengths in silylated alkoxyamines by
computational methods. Alkoxyamines 27 30 were chosen
as model compounds to elucidate the effect of N-silyl substi-
tution in alkoxyamines.
Structures were optimized with Gaussian98 (A.9)[21] at
the B3LYP level of theory. Unless otherwise stated, the
bond energies were BSSE (basis-superposition error)-cor-
worked well. P(n-BA)-b-PS-b-(n-BA) with
a
Mn of
30900 gmolꢀ1 and a narrow PDI of 1.14 was formed
1160
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 1156 1166

Steric and Electronic Effects in Cyclic Alkoxyamines
1156 1166
ꢀ
ꢁ
rected (counterpoise method).[22] Calculated C O bond en-
ergies and bond lengths of the alkoxyamines 27 30 are pre-
sented in Table 4.
ꢀ
½nitroxideŠ1ꢀ½nitroxideŠ
t
ln
¼ ꢀkd Á t
ð1Þ
½nitroxideŠ
1
ꢀ
ꢀ
The fastest C O bond homolysis was measured for the
Similar C O bond lengths were calculated for the silylat-
ed and N,N-tert-butylated alkoxyamines (compare 27 with
six-membered ethyl-substituted alkoxyamines 13 and 14
(Table 5; entries 1 and 2; 13: 2.2î10ꢀ2 sꢀ1; 14: 2.9î10ꢀ2 sꢀ1).
For the corresponding methyl derivatives, homolytic cleav-
age rate constants kd of 5.7î
ꢀ
29 and 28 with 30). As expected for steric reasons, C O
bonds in alkoxyamines 28 and 30, which are derived from
10ꢀ4 sꢀ1 (18 at 407 K) and 3.3î
10ꢀ3 sꢀ1 (19 at 413 K) have
ꢀ
Table 4. Calculated C O bond energies and bond lengths.
DE[C O][kJmol
]
been reported.[7] This clearly
shows that increasing the size
of the a-N substituents in
cyclic alkoxyamines leads to an
increase in the rate constant
ꢀ1
ꢀ
ꢀ
ꢀ
ꢀ
ꢀ
Compound
N O[ä]
Si N[ä]
C N[ä]
C O[ä]
27
28
29
30
1.451
1.449
1.476
1.470
1.515
1.527
1.422
1.480
1.423
1.473
ꢀ138.25
ꢀ78.38
1.797
1.806
ꢀ170.91
ꢀ119.20
ꢀ
for the C O bond homolysis,
in agreement with recent re-
tert-butyl radicals, are longer than the corresponding
methyl-substituted alkoxyamines 27 and 29. As a direct con-
sequence of the steric repulsion and the additional stability
ports from the Fischer laboratory for other cyclic alkoxya-
mines.[9] Interestingly, replacement of two of the four methyl
groups in 18 by ethyl groups leads to a rate constant that
lies in the middle between the those of tetramethyl deriva-
tive 18 and the tetraethyl derivative 13 (!15, kd of 4.9î
10ꢀ3 sꢀ1, Table 5, entry 3). To our surprise, the ring-enlarged
seven-membered alkoxyamines 16 and 17 homolyze more
slowly than the corresponding six-membered compounds 13
and 14 (compare Table 5, entries 1 and 2with 4 and 5). This
is in contrast to the methyl series where the seven-mem-
bered alkoxyamines 20 and 21 homolyze faster than the six-
membered derivatives 18 and 19 (20: 2.0î10ꢀ3 sꢀ1 at 407 K;
21: 2.7î10ꢀ2 sꢀ1 at 407 K).[7]
ꢀ
of a tertiary radical compared to a primary radical, C O
bonds are stronger in the methyl-substituted alkoxyamines
27 and 29 compared to their tert-butyl derivatives 28 and 30
(59.9 kJmolꢀ1 for 27/28 and 51.7 kJmolꢀ1 for 29/30). As pre-
ꢀ
dicted from our experimental results, C O bonds for the si-
lylated alkoxyamines are stronger than for the correspond-
ing dialkylated alkoxyamines. Thus, replacement of both
tert-butyl groups in 27 with two trimethylsilyl groups (!29)
ꢀ
leads to an increase of the C O bond energy by
32.7 kJmolꢀ1. In analogy, for the sterically more hindered
tert-butyl compounds, a 40.8 kJmolꢀ1 stronger C O bond
ꢀ
was calculated for the silyl-substituted alkoxyamine 30.
Hence, we can conclude that, in contrast to previous calcula-
tions, replacement of the N substituents in alkoxyamines by
Thermal stability of the alkoxyamines: The thermal decom-
position of a polymeric alkoxyamine leading to a terminally
unsaturated polymer and the corresponding hydroxylamine
is an important side reaction in NMP. This process can occur
by transfer of a b-hydrogen atom from the transient polymer
radical to the nitroxide or by a non-radical direct ionic elim-
ination.[24b] We decided to study the thermal decomposition
of the various alkoxyamines to styrene and the correspond-
ing hydroxylamine. To this end, the alkoxyamine was dis-
solved in a NMR tube in perdeutero p-xylene (0.03 0.06m).
The degassed sample was heated to 398 K within the cavity
ꢀ
silyl groups leads to an increase in the C O bond energy
and, therefore, silylated alkoxyamines are not useful as reg-
ulators/initiators for the controlled free-radical polymeriza-
tion.
ꢀ
Kinetics of the C O bond homolysis–EPR experiments:
The kinetic experiments were conducted in tert-butylben-
zene at 403 K. Oxygen was used to scavenge the styryl radi-
cal and the concentration of the released nitroxide was
measured by EPR spectroscopy, as previously descri-
bed.[5,6h,k,7,23] The experimental cleavage rate constants kd
were calculated by means of Equation (1) (conversions up
to 30%). The activation energies Ea were estimated from
the rate constants, whereby A = 2.4î1014 sꢀ1.[5,6h,7]
1
of a H NMR spectrometer (500 MHz) and the decomposi-
tion was followed by monitoring the decrease of the alkoxy-
amine signals as well as the increase of the styrene resonan-
ces. The signal of the benzylic H atom at ꢁ4.7 was used to
estimate the alkoxyamine concentration. Similar experi-
ments have previously been described.[7,24] Spectra were re-
ꢀ
Table 5. Rate constants for the C O bond homolysis and decomposition rate constants of alkoxyamines 13 17 and EPR parameters of the corresponding
nitroxides.
[a]
[b]
[c]
Entry
Alkoxyamine
kd [sꢀ1
]
Ea [kJmolꢀ1
]
kdecomp [sꢀ1
]
Nitroxide
aN [G][d]
g
1
2
3
4
5
13
14
15
16
17
2.2î10ꢀ2
2.9î10ꢀ2
4.9î10ꢀ3
5.9î10ꢀ3
1.2î10ꢀ2
123.7
122.8
128.7
128.1
125.7
1.5î10ꢀ5
1.2î10ꢀ4
1.1î10ꢀ5
6.6î10ꢀ5
2.1î10ꢀ4
6a
8
6b
10
12
14.16
14.51
14.51
13.58
14.09
2.006
2.006
2.006
2.006
2.006
[a] Measured at 403 K. [b] Ea was calculated with A = 2.4î1014 sꢀ1; see ref. [6h]. Statistical errors 2 3 kJmolꢀ1. [c] Measured at 398 K. [d] EPR data
were recorded in tBuPh saturated with O2 and are given with an error of Æ0.14 G.
1161
Chem. Eur. J. 2004, 10, 1156 1166
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

A. Studer et al.
FULL PAPER
corded every 5 minutes. The decomposition rate constants
(kdecomp) for the various alkoxyamines were determined with
Equation (2), according to Fukuda×s work[24a] ([S] = styrene
concentration; [A] = alkoxyamine concentration) and are
summarized in Table 5.
bound alkoxyamine initiator is difficult to determine. Forma-
tion of AA and AB diblock and ABA triblock copolymers
with excellent control was achieved with alkoxyamine 13.
Furthermore, we showed that silylated alkoxyamines are
not suitable as regulators/initiators for controlled living radi-
cal polymerization. In contrast to previous literature reports,
ln ð½SŠ=½AŠ þ 1Þ ¼ kdecomp Á t
ð2Þ
ꢀ
the C O bonds in silylated alkoxyamines are stronger than
ꢀ
the C O bonds in analogous N,N-dialkylated alkoxyamines.
The hydroxy-substituted alkoxyamine 14 decomposes one
order of magnitude faster than the corresponding keto alk-
oxyamine 13 (Table 3, entries 1,2). The broader polydisper-
sity obtained in the 14-mediated styrene polymerization can
therefore be explained by the instability of the regulator. It
is important to note that the ethyl-substituted six-membered
alkoxyamines are less stable than the methyl congeners
(kdecomp (18) = 5.9î10ꢀ6 sꢀ1; kdecomp (19) = 1.3î10ꢀ5 sꢀ1).[7,24a]
Because of this instability, the best polymerization results
with 13 were achieved at lower temperatures (1058C). Al-
koxyamine 14 is not stable enough for an efficient regulator/
initiator. Similar trends were observed for the seven-mem-
bered alkoxyamines 16 and 17 (Table 5, entries 4 and 5).[25]
The hydroxy derivative 17 was the least stable compound
studied. Similar decomposition rate constants have been ob-
tained for the corresponding methyl substituted alkoxya-
mines 20 and 21 (see ref. [7]) In agreement with the experi-
mental results, the replacement of the a-N-methyl groups in
seven-membered alkoxyamines by ethyl groups has only a
small effect on the decomposition rate constant.
Experimental Section
General: H and 13C NMR spectra were recorded on a Bruker ARX500,
1
ARX400, ARX300, ARX200 or a AC200 spectrometer. Chemical shifts
are referenced to SiMe4 as an internal standard. TLC was carried out on
Merck silica gel60F254 plates, detection by UV or dipping into a solution
of KMnO4 (3.0 g), NaHCO3 (10.0 g), and H2O (800 mL) or a solution of
Ce(SO4)2¥H2O (10 g), phosphomolybdic acid hydrate (25 g), concentrated
H2SO4 (60 mL), and H2O (940 mL), followed by heating. FC was carried
out on Merck or Fluka silica gel60 (40 63 mm) at ꢁ0.4 bar. IR spectra
were recorded on a IR750 (Nicolet Magna) or a IFS-200 (Bruker). MS
were recorded on a VGTribid, VarianCH7 (EI), IonSpec Ultima, Finni-
gan MATTSQ700 or a Finnigan MAT95S (ESI) in m/z (% of basis
peak). Size-exclusion chromatography (SEC) was carried out with THF
as eluent at a flow rate of 1.0 mLminꢀ1 at room temperature on a system
consisting of a L6200A Intelligent Pump (Merck Hitachi), a set of two
PLgel 5 mm MIXED-C columns (300î7.5 mm, Polymer Laboratories),
and a RI-101 detector (Shodex). Data were acquired through a PL Data-
stream unit (Polymer Laboratories) and analyzed with Cirrus GPC soft-
ware (Polymer Laboratories) based upon calibration curves based on pol-
ystyrene and poly(methyl methacrylate) standards (Polymer Laboratories
Polystyrene Medium MW Calibration Kit S-M-10 to determine the mo-
lecular weight of polystyrene and Polymer Laboratories Polymethylme-
thacrylate Medium MW Calibration Kit M-M-10 to determine the molec-
ular weight of n-butyl acrylate) with peak molecular weights ranging
from 500 3000000 gmolꢀ1. EPR spectra were recorded on a ESP300E
(Bruker) equipped with a Nicolet Cavity (Bruker) and a B-TC 80/15
(Bruker). The nitroxide concentrations were determined by double inte-
gration of the EPR spectra and calibration with a TEMPO solution in
tert-butylbenzene. Microwave-assisted heating was performed in an MLS-
Ethos1600 Microwave System (MLS). Solvents were purified by standard
methods. Compounds sensitive to air and moisture were handled under
argon by means of Schlenk techniques.
Conclusion
We presented the synthesis of new sterically hindered six-
and seven-membered cyclic alkoxyamines. The alkoxya-
mines were tested as regulator/initiators in controlled/living
radical polymerizations. We showed that replacement of
methyl groups in the a position to the N atom in six-mem-
bered cyclic alkoxyamines by ethyl groups leads to highly ef-
ficient regulators. For the seven-membered cyclic alkoxya-
mines, however, the replacement of methyl by ethyl groups
only has a small (detrimental) effect on the polymerization.
We presented kinetic data that show that the replacement of
methyl by ethyl groups in six-membered cyclic alkoxyamines
(4-Ethyl-2-oxohex-3-enyl)-dimethylphosphonate (2): A solution of diiso-
propylamine (5.6 mL, 40.13 mmol) in THF (250 mL) was treated with bu-
tyllithium (BuLi) in hexanes (40.13 mmol) at 08C. After stirring for
30 min at 08C, ketodiphosphonate 1 (10.00 g, 36.48 mmol) dissolved in
THF (50 mL) was added dropwise. The solution turned deep yellow.
After stirring for another 40 min, the solution was chilled to ꢀ358C and
BuLi in hexanes (80.26 mmol) was added dropwise. The reaction mixture
turned dark red. After stirring for 1 h at ꢀ358C, 3-pentanone was added
slowly. The solution was allowed to warm to room temperature and was
stirred overnight. The reaction mixture was treated with brine, extracted
with t-butyl methyl ether (MTBE), and the organic layer was dried
(MgSO4). Evaporation of the solvents in vacuo yielded the crude product
2 (8.47 g), which was used for the next step without further purification.
An analytical sample was purified by FC (MTBE/acetone 4:1 to acetone/
ꢀ
leads to an increase of the rate constant of the C O bond
homolysis, which is an important parameter in NMP. Along
with the desired increase of the homolysis rate constant, a
decrease of the alkoxyamine stability was measured for the
ethyl derivatives. However, we found that the six-membered
alkoxyamine 13 is sufficiently stable and sufficiently active
at 1058C. Controlled polymerization of styrene and n-butyl
acrylate was achieved by the use of this new sterically highly
hindered alkoxyamine. Controlled polymerizations can even
be performed at 908C. It is important to note that NMP of
n-butyl acrylate is currently only possible if noncyclic al-
koxyamines are used.[1c,19] Furthermore, addition of sacrifi-
cial nitroxide is not necessary for the polymerization of n-
butyl acrylate with our new initiator/regulator 13. This is im-
portant for the formation of polymer brushes from surfaces
(grafting from) where the exact concentration of surface-
MeOH 10:1). IR (film): n˜
= 3468br, 2970m, 1684s, 1613s, 1463m,
1
1396w, 1258s, 1032s cmꢀ1; H NMR (200 MHz, CDCl3): d = 6.11 (s, 1H,
CH), 3.76 (d, J(H,P) = 11.2Hz, 6H, CH 3), 3.08 (d, J(H,P) = 22.5 Hz,
2H, CH2), 2.55 (q, J = 7.5 Hz, 2H, CH2), 2.19 (dq, J1 = 7.5 Hz, J2
1.3 Hz, 2H, CH2), 1.06 (t, J 7.5 Hz, 3H, CH3), 1.02ppm (t,
=
=
=
J
7.5 Hz, 3H, CH3); 13C NMR (75 MHz, CDCl3): d = 190.2(d, J(C,P) =
6.2Hz, C), 169.4 (C), 121.0 (d, J(C,P) = 2.2 Hz, CH), 52.7 (d, J(C,P) =
6.2Hz, 2CH 3), 42.5 (d, J(C,P) = 126.9 Hz, CH2), 31.1 (CH2), 25.9 (CH2),
12.5 (CH3), 11.8 ppm (CH3); MS (EI): 234 (25 [M]+), 151 (44), 124 (100),
111 (26), 95 (70), 55 (29); HRMS (EI) calcd for C10H19O5P ([M]+):
234.1021; found: 234.1019.
1162
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 1156 1166

Steric and Electronic Effects in Cyclic Alkoxyamines
1156 1166
3,7-Diethylnona-3,6-dien-5-one (3a) and 3,7-diethylnona-2,6-dien-5-one
(4): To the phosphonate 2 (8.47 g, 36.15 mmol) dissolved in benzene
(90 mL) was added CsOH¥H2O (15.18 g, 90.37 mmol), 3-pentanone
(38.3 mL, 0.362mol) and water (2.3 mL). After heating to 60 8C for 14 h
and then refluxing for 4 h, the reaction mixture was cooled to room tem-
perature and treated with a 2m HCl solution. The mixture was then ex-
tracted with diethyl ether, and the organic layer was dried (MgSO4).
Evaporation of the solvents in vacuo and purification of the crude prod-
uct by FC (diethyl ether/pentane 1:100) afforded the unsaturated ketones
3a and 4 (2.21 g, 3/4 1:6.4, 31% over two steps) as a yellow oil. The un-
desired isomer 4 could be transformed into the desired isomer 3a under
(200 MHz, CDCl3): d = 2.22 (s, 2H, CH2), 2.18 (s, 2H, CH2), 1.57 1.24
(m, 4H, CH2), 1.18 (s, 6H, CH3), 0.77 ppm (t, J = 7.5 Hz, 6H, CH3); 13C
NMR (50 MHz, CDCl3): d = 211.4 (C), 59.7 (C), 54.5 (C), 54.4 (CH2),
50.4 (CH2), 32.1 (CH3), 32.0 (CH2), 7.6 ppm (CH3); MS (ESI): 367 (14
[2M+H]+), 184 (100, [M+H]+); HRMS (ESI) calcd for C11H22NO
([M+H]+): 184.1701; found: 184.1708.
To
a
solution of 2,2-diethyl-5,5-dimethylpiperidin-4-one (271 mg,
1.48 mmol) in MeOH (3.0 mL) and water (1.0 mL) was added
Na2WO4¥2H2O (81 mg, 0.25 mmol) and a 35% aqueous solution of H2O2
(0.90 mL, 8.87 mmol). The reaction mixture was stirred for 24 h at 258C,
quenched with brine and extracted with MTBE. The organic layer was
dried (MgSO4). Evaporation of the solvents in vacuo yielded the 6b
(85 mg, 29%) as an orange oil. EPR: aN = 14.51 G.
basic conditions. IR (film): n˜
= 3440br, 2969s, 2935s, 2875s, 1670m,
1621s, 1426m, 1119m cmꢀ1 1H NMR (300 MHz, CDCl3): d = 6.00 (s,
;
2H, CH), 2.60 (q, J = 7.5 Hz, 4H, CH2), 2.17 (q, J = 7.3 Hz, 4H, CH2),
1.07 ppm (t, J = 7.3 Hz, 12H, CH3); 13C NMR (50 MHz, CDCl3): d =
191.4 (C), 164.8 (C), 164.7 (C), 124.0 (2CH), 30.9 (CH2), 25.5 (CH2), 13.0
(CH3), 12.1 ppm (CH3); MS (EI): 194 (10, [M]+), 165 (69), 137 (33), 111
(100), 55 (77); HRMS (EI) calcd for C13H22O ([M]+): 194.1671; found:
194.1677.
2,2,6,6-Tetraethylpiperidin-4-ol (7): To
a solution of aminoketone 5
(200 mg, 0.95 mmol) in THF (10 mL) was added lithium aluminum hy-
dride (LAH) (36 mg, 0.95 mmol). The reaction mixture was stirred at
258C for 12h, then heated to reflux for 4 h. After the mixture had been
cooled to room temperature, water (45 mL), a 15% aqueous solution of
NaOH (45 mL), and water (90 mL) were added to the stirred mixture in
5 min intervals. The suspension was finally allowed to stir for 20 min
while a white precipitate formed. Filtration, washing with MTBE, drying
(MgSO4), and evaporation of the solvents in vacuo yielded alcohol 7
(203 mg, 98%) which was used without further purification. IR (film):
6-Ethyl-2-methylocta-2,5-dien-4-one (3b): To a solution of diisopropyl-
amine (2.7 mL, 19.13 mmol) in THF (95 mL) was added BuLi in hexanes
(19.13 mmol) at 08C. After the mixture had been stirred for 15 min, a sol-
ution of phosphonate 2 (2.99 g, 12.76 mmol) in THF (45 mL) was added
dropwise. The mixture was heated to reflux for 30 min, then acetone
(7.26 mL, 127.6 mmol) was added. After refluxing overnight, the reaction
mixture was cooled to room temperature and a saturated aqueous solu-
tion of NH4Cl was added. Extraction (diethyl ether), drying (MgSO4),
evaporation of the solvents in vacuo, and purification by FC (diethyl
ether/pentane 1:50) yielded 3b (484 mg, 22%). IR (film): n˜ = 3441br,
n˜ = 3343br, 2965s, 2965s, 2876m, 1460s, 1378m, 1058s, 1031m cmꢀ1 1H
;
NMR (300 MHz, CDCl3): d = 3.95 (tt, J1 = 11.6 Hz, J2 = 4.0 Hz, 2H,
CH2), 1.84 (dd, J1 = 11.9 Hz, J2 = 3.6 Hz, 2H, CH2), 1.57 1.16 (m, 8H,
CH2), 0.93 (dd, J1 = J2 = 11.9 Hz, 2H, CH2), 0.78 ppm (dt, J1 = 7.3 Hz,
J2 = 3.0 Hz, 12H, CH3); 13C NMR (75 MHz, CDCl3): d = 64.4 (CH),
55.8 (2C), 44.4 (2CH 2), 34.5 (2CH2), 29.9 (2CH2), 8.4 (2CH3), 7.5 ppm
(2CH3); MS (ESI): 214 (100, [M+H]+), 86 (4); HRMS (ESI) calcd for
C13H28NO ([M+H]+): 214.2171; found: 214.2177.
2970s, 2935s, 2876s, 1673m, 1627s, 1443m, 1114m cmꢀ1 1H NMR
;
(300 MHz, CDCl3): d = 6.06 (s, 1H, CH), 5.95 (s, 1H, CH), 2.57 (q, J =
7.3 Hz, 2H, CH2), 2.18 2.11 (m, 5H, CH2, CH3), 1.86 (s, 3H, CH3), 1.05
(t, J = 7.3 Hz, 3H, CH3), 1.04 ppm (t, J = 7.3 Hz, 3H, CH3); 13C NMR
(75 MHz, CDCl3): d = 191.4 (C), 165.2 (C), 154.0 (C), 126.3 (CH), 123.6
(CH), 30.9 (CH2), 27.6 (CH3), 25.5 (CH2), 20.4 (CH3), 13.0 (CH3),
12.1 ppm (CH3); MS (EI): 166 (32, [M]+), 151 (100), 123 (48), 109 (26),
83 (75), 55 (47); HRMS (EI) calcd for C11H18O ([M]+): 166.1358; found:
166.1357.
2,2,6,6-Tetraethylpiperidin-4-ol-N-oxyl radical (8): To a solution of the
aminoketone
7 (579 mg, 2.71 mmol) in MeOH (5.5 mL) and water
(1.8 mL) was added Na2WO4¥2H2O (149 mg, 0.45 mmol) and a 35%
aqueous solution of H2O2 (1.63 mL, 16.26 mmol). The reaction mixture
was stirred for 24 h at 258C, quenched with brine, and extracted with
MTBE. The organic layer was dried (MgSO4). Evaporation of the sol-
vents in vacuo yielded nitroxide 8 (625 mg, 97%) as an orange oil. EPR:
aN = 14.51 G.
2,2,6,6-Tetraethylpiperidin-4-one (5): The ketones 3a and
4 (2.21 g,
11.43 mmol) and a concentrated aqueous solution of NH4OH (17.6 mL,
114.3 mmol) were heated to 1058C in a sealed tube for 20 h. The reaction
mixture was cooled to room temperature, treated with brine, and extract-
ed with MTBE. The organic layer was dried (MgSO4). Evaporation of
the solvents in vacuo and purification of the crude product by FC (dieth-
yl ether/pentane 1:5) afforded the piperdinone 5 (960 mg, 40%) as an
orange oil along with unreacted 3a and 4 (700 mg, 32%). IR (film): n˜ =
2,2,7,7-Tetraethylazepan-4-one (9): To
a solution of aminoketone 5
(620 mg, 2.93 mmol) in CH2Cl2 (30 mL) over 3 ä molecular sieves (8.8 g)
at ꢀ788C were added BF3¥OEt2 (0.40 mL, 3.22 mmol) and subsequently a
2m solution of Me3SiCHN2 in Et2O (4.4 mL, 8.79 mmol). The reaction
mixture was stirred at ꢀ788C for 3.5 h and then quenched with a saturat-
ed aqueous solution of NaHCO3 at ꢀ788C. Filtration, extraction with di-
ethyl ether, drying of the organic layers (MgSO4), and evaporation of the
solvents in vacuo yielded the crude homologue. This was dissolved in
MeOH (50 mL), and pyridinium p-toluenesulfonate (92mg, 0.37 mmol)
was added. After the mixture had been stirred for 14 h at 258C, Et3N
(0.21 mL, 1.48 mmol) was added. Then the mixture was stirred for an ad-
ditional 30 min, the solvent was evaporated in vacuo, and a saturated solu-
tion of NH4Cl was added. Extraction (MTBE), drying (MgSO4), evapora-
tion of the solvents in vacuo, and purification by FC (diethyl ether/pen-
tane 1:6) yielded the azepane 9 (521 mg, 79%). IR (film): n˜ = 3453w,
1
3354br, 2945s, 2832s, 2524w, 2045w, 1451m, 1032s, 662br cmꢀ1; H NMR
(200 MHz, CDCl3): d
= 2.18 (s, 4H, CH2), 1.47 1.28 (m, 8H, CH2),
0.77 ppm (t, J = 5.1 Hz, 12H, CH3); 13C NMR (50 MHz, CDCl3): d =
212.1 (C), 58.4 (2C), 49.7 (2CH2), 33.0 (4CH2), 7.9 ppm (4CH3); MS
(ESI): 234 (60 [M+Na]+), 212 (13, [M+H]+), 194 (25), 154 (100), 111
(28); HRMS (ESI) calcd for C13H25NNaO ([M+Na]+): 234.1834; found:
234.1840.
2,2,6,6-Tetraethylpiperidin-4-one-N-oxyl radical (6a): To a solution of the
aminoketone
5 (300 mg, 1.42mmol) in MeOH (3.0 mL) and water
2962s, 2930s, 1706s, 1456m, 1359m, 1271w, 1176w cmꢀ1 1H NMR
;
(1.0 mL) was added Na2WO4¥2H2O (118 mg, 0.36 mmol) and a 35%
aqueous solution of H2O2 (0.85 mL, 8.52mmol). The reaction mixture
was stirred for 24 h at 258C, quenched with a saturated aqueous solution
of K2CO3 and extracted with diethyl ether. The organic layer was dried
(MgSO4). Evaporation of the solvents in vacuo and purification by FC
(diethyl ether/pentane 1:3) yielded 6a (280 mg, 87%) as an orange oil.
EPR: aN = 14.16 G.
(400 MHz, CDCl3): d = 2.56 (s, 2H, CH2), 2.30 (t, J = 6.2Hz, 2H,
CH2), 1.86 (t, J = 6.2Hz, 2H, CH 2), 1.51 1.45 (m, 4H, CH2), 1.39 (q,
J
=
7.5 Hz, 4H, CH2), 0.85 0.77 ppm (m, 12H, CH3); 13C NMR
(100 MHz, CDCl3): d = 212.6 (C), 57.4 (C), 57.3 (C), 51.1 (CH2), 38.9
(CH2), 32.4 (2CH2), 32.0 (CH2), 31.0 (2CH2), 8.0 (2CH3), 7.9 ppm
(2CH3); MS (ESI): 226 (100, [M+H]+); HRMS (ESI) calcd for
C14H28NO ([M+H]+): 226.2171; found: 226.2165.
2,2-Diethyl-6,6-dimethylpiperidin-4-one-N-oxyl radical (6b): 6-Ethyl-2-
methylocta-2,5-dien-4-one 3b (250 mg, 1.29 mmol) and a concentrated
aqueous solution of NH4OH (2.0 mL, 12.87 mmol) were heated to 1058C
in a sealed tube for 17 h. After cooling to room temperature, the reaction
mixture was treated with brine and then extracted with MTBE. The or-
ganic layer was dried (MgSO4). Evaporation of the solvents in vacuo and
purification of the crude product by FC (diethyl ether/pentane 1:5) af-
forded 2,2-diethyl-5,5-dimethylpiperidin-4-one (261 mg, 92%). IR (film):
2,2,7,7-Tetraethylazepan-4-one-N-oxyl radical (10): To a solution of the
azepanone 9 (457 mg, 2.03 mmol) in MeOH (4.8 mL) and water (1.4 mL)
was added Na2WO4¥2H2O (112mg, 0.34 mmol) and a 35% aqueous solu-
tion of H2O2 (1.22 mL, 12.2 mmol). The reaction mixture was stirred for
18 h at 258C, quenched with brine, and extracted with diethyl ether. The
organic layer was dried (MgSO4). Evaporation of the solvents in vacuo
and purification by FC (diethyl ether/pentane 1:8) yielded 10 (200 mg,
41%) as an orange oil. EPR: aN = 13.58 G.
1
n˜ = 3333br, 2966s, 1709s, 1620w, 1462m, 1381m, 1295m cmꢀ1; H NMR
1163
Chem. Eur. J. 2004, 10, 1156 1166
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

A. Studer et al.
FULL PAPER
2,2,7,7-Tetraethylazepan-4-ol (11): To a solution of azepanone 9 (63 mg,
0.32mmol) in THF (2mL) was added LAH (36 mg, 0.32mmol). The re-
action mixture was stirred at 258C for 12h, then heated to reflux for 4 h.
After the mixture had been cooled to room temperature, water (15 mL),
a 15% aqueous solution of NaOH (15 mL), and water (30 mL) were
added with 5 min intervals with stirring. The suspension was finally al-
lowed to stir for 20 min while a white precipitate formed. Filtration,
washing with MTBE, drying (MgSO4), and evaporation of the solvents in
vacuo yielded the alcohol 11 (57 mg, 78%), which was used without fur-
ther purification. IR (film): n˜ = 3347br, 2966s, 2936s, 2878m, 1461s,
1305m cmꢀ1
;
1H NMR (300 MHz, CDCl3): d = 7.35 7.23 (m, 5H, CH),
4.77 (br, 1H, CH), 2.41 2.27 (m, 4H, CH2), 2.32 (s, 3H, CH3), 2.04 (s,
3H, CH3), 1.49 (d, J = 5.4 Hz, 3H, CH3), 1.31 1.21 (m, 10H), 0.88 ppm
(t, J = 5.6 Hz, 6H, CH3); 13C NMR (75 MHz, CDCl3): d = 209.8 (C),
144.2 (C), 128.1 (2CH), 127.4 (CH), 126.6 (2CH), 83.3 (CH), 65.8 (C),
60.4 (C), 53.1 (CH2), 47.4 (CH2), 33.4 (CH3), 33.3 (CH3), 30.9 (CH2), 28.5
(CH2), 23.5 (CH3), 9.5 (CH3), 8.4 ppm (CH3); MS (ESI): 326 (100,
[M+Na]+), 304 (41, [M+H]+), 242 (53); HRMS (ESI) calcd for
C19H29NNaO2 ([M+Na]+): 326.2096; found: 326.2091.
2,2,7,7-Tetraethyl-1-(1-phenylethoxy)azepan-4-one (16): Nitroxide 10
(200 mg, 0.83 mmol), Cu dust (53 mg, 0.83 mmol), Cu(OTf)2 (15 mg,
0.042mmol), (4,4 ’-di-tert-butyl)-2,2’-bipyridine (23 mg, 0.168 mmol), 1-
phenylethylbromide (154 mg, 0.83 mmol), and benzene (3.0 mL) were
heated for 48 h to 758C in a sealed tube under an argon atmosphere. Fil-
tration over a thin pad of SiO2, washing with MTBE, evaporation of the
solvents in vacuo, and purification by FC (diethyl ether/pentane 1:30)
yielded 16 (64 mg, 22%) as a mixture of two diastereoisomers because of
the chirality at the nitrogen atom (d.r. = 1.1:1). IR (film): n˜ = 3369w,
2963s, 2878s, 1705s, 1462s, 1359m, 1180m cmꢀ1; MS (ESI): 346 (36,
[M+H]+), 305 (100); HRMS (ESI) calcd for C22H36NO2 ([M+H]+):
346.2746; found: 346.2739.
1377m, 11180m, 1032m cmꢀ1 1H NMR (400 MHz, CDCl3): d = 3.95
;
3.89 (m, 1H, CH), 1.86 1.76 (m, 2H, CH2), 1.70 1.57 (m, 3H, CH2),
1.55 1.24 (m, 9H, CH2), 0.83 0.72ppm (m, 12H, CH 3); 13C NMR
(100 MHz, CDCl3): d = 69.6 (CH), 57.3 (C), 56.5 (C), 45.7 (CH2), 34.0
(CH2), 33.5 (CH2), 32.5 (CH2), 32.4 (CH2), 30.8 (CH2), 30.0 (CH2), 8.5
(CH3), 8.2(CH 3), 7.8 (CH3), 7.2ppm (CH 3); MS (ESI): 250 (12,
[M+Na]+), 228 (100, [M+H]+); HRMS (ESI) calcd for C14H30NO
([M+H]+): 228.2327; found: 228.2334.
2,2,7,7-Tetraethylazepan-4-ol-N-oxyl radical (12): To a solution of the
azepanol 11 (151 mg, 0.66 mmol) in MeOH (1.5 mL) and water (0.5 mL)
was added Na2WO4¥2H2O (36 mg, 0.11 mmol) and a 35% aqueous solu-
tion of H2O2 (0.40 mL, 3.96 mmol). The reaction mixture was stirred for
24 h at 258C, quenched with brine, and extracted with MTBE. The organ-
ic layer was dried (MgSO4). Evaporation of the solvents in vacuo yielded
12 (170 mg, 98%) as an orange oil. EPR: aN = 14.09 G.
1
Major isomer: H NMR (500 MHz, CDCl3): d = 7.29 7.25 (m, 4H, CH),
7.23 7.20 (m, 1H, CH), 4.54 (q, J = 6.9 Hz, 1H, CH), 2.79 (dd, J1
=
12.4 Hz, J2 = 7.1 Hz, 2H, CH2), 2.29 0.80 (m, 10H, CH2), 2.14 (s, 2H,
CH2), 1.39 (d, J = 6.9 Hz, 3H, CH3), 1.26 (t, J = 7.1 Hz, 3H, CH3), 0.92
(t, J = 7.6 Hz, 3H, CH3), 0.84 (t, J = 7.6 Hz, 3H, CH3), 0.71 ppm (t, J
= 7.6 Hz, 3H, CH3); 13C NMR (125 MHz, CDCl3): d = 208.5 (C), 145.0
(C), 128.0 (2CH), 127.4 (CH), 126.9 (2CH), 82.2 (CH), 71.3 (C), 65.1
(C), 50.2(CH 2), 32.3 (CH2), 31.5 (CH2), 31.1 (CH2), 25.7 (CH2), 24.4
(CH2), 22.7 (CH3), 10.2(CH ), 9.2(CH ), 8.6 (CH3), 8.2ppm (CH 3).
2,2,6,6-Tetraethyl-1-(1-phenylethoxy)piperidin-4-one (13): Nitroxide 6a
(510 mg, 2.23 mmol), Cu dust (142 mg, 2.23 mmol), Cu(OTf)2 (40 mg,
0.11 mmol), (4,4’-di-tert-butyl)-2,2’-bipyridine (60 mg, 0.45 mmol), 1-phe-
nylethylbromide (413 mg, 2.23 mmol), and benzene (7.0 mL) were heated
for 18 h to 758C in a sealed tube under an argon atmosphere. Filtration
over a thin pad of SiO2, washing with MTBE, evaporation of the solvents
in vacuo, and purification by FC (diethyl ether/pentane 1:20) yielded 13
3
3
1
Minor isomer: H NMR (500 MHz, CDCl3): d = 7.29 7.25 (m, 4H, CH),
7.23 7.20 (m, 1H, CH), 4.59 (q, J = 6.7 Hz, 1H, CH), 2.76 (dd, J1
=
(456 mg, 61%). IR (film): n˜
= 3029w, 2971s, 2881m, 1718s, 1464m,
1
1251m, 1060s cmꢀ1; H NMR (400 MHz, CDCl3): d = 7.32 7.24 (m, 5H,
CH), 4.72(q, J = 6.7 Hz, 1H, CH), 2.46 2.26 (brm, 4H, CH2), 2.13 1.98
(brm, 1H, CH2), 1.89 1.55 (brm, 7H, CH2), 1.46 (d, J = 6.7 Hz, 3H,
CH3), 1.15 0.57 ppm (brm, 12H, CH3); 13C NMR (100 MHz, CDCl3): d
12.4 Hz, J2 = 6.9 Hz, 2H, CH2), 2.29 0.80 (m, 10H, CH2), 2.07 (s, 2H,
CH2), 1.38 (d, J = 6.7 Hz, 3H, CH3), 0.95 (t, J = 7.3 Hz, 3H, CH3), 0.92
(t, J = 7.3 Hz, 3H, CH3), 0.67 (t, J = 7.3 Hz, 3H, CH3), 0.66 ppm (t, J
= 7.3 Hz, 3H, CH3); 13C NMR (125 MHz, CDCl3): d = 208.5 (C), 144.9
(C), 128.0 (2CH), 127.2 (CH), 126.6 (2CH), 82.1 (CH), 71.0 (C), 65.9
(C), 50.0 (CH2), 32.2 (CH2), 31.8 (CH2), 31.2(CH 2), 25.4 (CH2), 24.3
(CH2), 23.4 (CH3), 10.0 (CH3), 9.4 (CH3), 8.7 (CH3), 7.3 ppm (CH3).
=
211.0 (C), 145.0 (C), 128.1 (2CH), 127.2 (CH), 126.5 (2CH), 83.1
(CH), 66.2(C), 66.1 (C), 46.7 (CH 2), 31.3 (CH2), 31.0 (CH2), 29.3 (CH2),
29.0 (CH2), 23.5 (CH3), 9.9 (CH3), 9.5 (CH3), 8.6 (CH3), 8.3 ppm (CH3);
MS (ESI): 354 (31, [M+Na]+), 260 (100); HRMS (ESI) calcd for
C21H33NNaO2 ([M+Na]+): 354.2409; found: 354.2393.
2,2,7,7-Tetraethyl-1-(1-phenylethoxy)azepan-4-ol (17): Nitroxide 12
(81 mg, 0.33 mmol), Cu dust (20 mg, 0.32 mmol), Cu(OTf)2 (5.4 mg,
0.015 mmol), (4,4’-di-tert-butyl)-2,2’-bipyridine (8.1 mg, 0.060 mmol), 1-
phenylethylbromide (56 mg, 0.30 mmol), and benzene (1.0 mL) were
heated for 14 h to 758C in a sealed tube under an argon atmosphere. Fil-
tration over a thin pad of SiO2, washing with MTBE, evaporation of the
solvents in vacuo, and purification by FC (MTBE/pentane 1:2) yielded
the alkoxyamine 17 (76 mg, 73%) as a mixture of diastereoisomers
(d.r. = 1.7:1). IR (film): n˜ = 3447br, 2972s, 2879s, 1691w, 1465s, 1373m,
2,2,6,6-Tetraethyl-1-(1-phenylethoxy)piperidin-4-ol (14): Nitroxide
8
(200 mg, 0.88 mmol), Cu dust (54 mg, 0.85 mmol), Cu(OTf)2 (16 mg,
0.044 mmol), (4,4’-di-tert-butyl)-2,2’-bipyridine (24 mg, 0.176 mmol), 1-
phenylethylbromide (154 mg, 0.83 mmol), and benzene (3.0 mL) were
heated for 19 h to 758C in a sealed tube under an argon atmosphere. Fil-
tration over a thin pad of SiO2, washing with MTBE, evaporation of the
solvents in vacuo, and purification by FC (diethyl ether/pentane 1:5)
yielded 14 (83 mg, 30%). IR (film): n˜ = 3440br, 2970s, 2880m, 1727w,
1056s cmꢀ1 13C NMR (100 MHz, CDCl3): d = 146.4 (C), 127.8 (2CH),
;
1537w, 1463m, 1375m cmꢀ1
;
1H NMR (400 MHz, CDCl3): d = 7.33 7.19
126.6 (CH), 125.8 (2CH), 82.3 (CH), 65.4 (C), 65.1 (C), 62.6 (CH), 40.0
(CH2), 39.6 (CH2), 30.1 (CH2), 29.6 (CH2), 29.1 (CH2), 27.3 (CH2), 27.0
(CH2), 24.8 (CH3), 10.1 (CH3), 9.8 (CH3), 8.2(CH 3), 7.9 ppm (CH3); MS
(ESI): 370 (25, [M+Na]+), 348 (19, [M+H]+), 318 (100); HRMS (ESI)
calcd for C22H37NNaO2 ([M+H]+): 370.2722; found: 370.2709.
(m, 5H, CH), 4.68 (q, J = 6.7 Hz, 1H, CH), 3.88 (tt, J1 = 11.4 Hz, J2
=
4.0 Hz, 1H, CH), 2.20 2.10 (m, 1H, CH2), 2.01 1.92 (m, 1H, CH2), 1.81
1.60 (m, 5H, CH2), 1.55 1.24 (m, 3H, CH2), 1.41 (d, J = 6.7 Hz, 3H,
CH3), 1.07 (t, J = 7.3 Hz, 3H, CH3), 0.97 0.75 (m, 2H, CH2), 0.92(t,
J
= 7.7 Hz, 3H, CH3), 0.70 (t, J = 7.5 Hz, 3H, CH3), 0.64 ppm (t, J =
7.1 Hz, 3H, CH3); 13C NMR (100 MHz, CDCl3): d = 146.5 (C), 127.9
(2CH), 126.6 (CH), 125.9 (2CH), 82.3 (CH), 65.5 (C), 65.1 (C), 62.7
(CH), 40.0 (CH2), 39.6 (CH2), 30.1 (CH2), 29.1 (CH2), 27.4 (CH2), 27.0
(CH2), 24.9 (CH3), 10.2(CH 3), 9.9 (CH3), 8.2(CH 3), 8.0 ppm (CH3); MS
(ESI): 334 (100, [M+H]+), 318 (86); HRMS (ESI) calcd for C21H36NO2
([M+H]+): 334.2746; found: 334.2740.
1
Major isomer: H NMR (500 MHz, CDCl3): d = 7.30 7.22 (m, 5H, CH),
4.69 4.66 (m, 1H, CH), 3.96 3.88 (m, 1H, CH), 2.34 0.53 ppm (m, 29H).
1
Minor isomer: H NMR (500 MHz, CDCl3): d = 7.30 7.22 (m, 5H, CH),
4.76 4.71 (m, 1H, CH), 4.11 4.05 (m, 1H, CH), 2.34 0.53 ppm (m, 29H).
1,2-Bis(chlorodiethylsilanyl)benzene (23): 1,2-Dibromobenzene (6.58 g,
27.90 mmol) was added dropwise to magnesium turnings (1.40 g,
57.68 mmol) and diethyl chlorosilane (7.12g, 58.01 mmol) in THF
(35 mL). The mixture was heated to reflux for 4 h, and, after cooling to
room temperature, hexane was added (30 mL). The organic layer was
washed with 2n HCl (30 mL) and water (30 mL) and was dried (MgSO4).
Evaporation of the solvents in vacuo and purification by distillation (ꢁ
908C, 1.1 mbar) yielded the bis(diethylsilanyl)benzene (2.04 g, 29%); 1H
NMR (300 MHz): d = 7.83 (dd, J1 = 5.4 Hz, J2 = 3.3 Hz, 2H, CH), 7.60
(dd, J1 = 5.6 Hz, J2 = 3.1 Hz, 2H, CH), 4.37 (qn, J = 3.3 Hz, 2H, SiH),
0.97 0.91 (m, 12H, CH3), 0.90 0.88 ppm (m, 8H, CH2).
2,2-Diethyl-6,6-dimethyl-1-(1-phenylethoxy)piperidin-4-one (15): 2,2-Di-
ethyl-6,6-dimethylpiperidin-4-one-N-oxyl
radical
(6b)
(72mg,
0.363 mmol), Cu dust (22 mg, 0.347 mmol), Cu(OTf)2 (6 mg, 0.017 mmol),
(4,4’-di-tert-butyl)-2,2’-bipyridine (18 mg, 0.066 mmol), 1-phenylethylbro-
mide (61 mg, 0.320 mmol), and benzene (1.2 mL) were heated for 24 h to
758C in a sealed tube under an argon atmosphere. Filtration over a thin
pad of SiO2, washing with MTBE, evaporation of the solvents in vacuo,
and purification by FC (diethyl ether/pentane 1:10) yielded 15 (79 mg,
72%). IR (film): n˜ = 3441w, 2974s, 2938s, 2880m, 1719s, 1494w, 1454m,
1164
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 1156 1166

Steric and Electronic Effects in Cyclic Alkoxyamines
1156 1166
Bis(diethylsilanyl)benzene (2.03 g, 8.10 mmol) was saturated with chlor-
an aluminum dish. Residual monomer was removed in a vacuum drying
cabinet at 608C for 12h. Conversion was evaluated gravimetrically; mo-
lecular weight and polydispersity index (PDI) were determined by size-
ine at 08C. Purging with argon removed excess chlorine and yielded the
product 23 (2.34 g, 90%); 1H NMR (300 MHz): d
= 7.86 (dd, J1 =
exclusion chromatography (SEC). Conversion
=
69%; Mn
=
5.6 Hz, J2 = 3.4 Hz, 2H, CH), 7.44 (dd, J1 = 5.6 Hz, J2 = 3.2Hz, 2H,
CH), 1.25 1.15 (m, 8H, CH2), 1.06 0.99 ppm (m, 12H, CH3). The physi-
cal data are in agreement with the values reported in literature.[15,16]
9400 gmolꢀ1; PDI = 1.09.
Typical procedure for the polymerization of n-butyl acrylate: A Schlenk
tube was charged with the alkoxyamine initiator 13 (30 mg, 90.0 mmol)
and n-butyl acrylate (1.28 mL, 8.996 mmol) under argon. The n-butyl ac-
rylate was previously degassed in three freeze thaw cycles and sealed off
under argon. The polymerization was carried out under argon at 1058C
for 32h. The resulting mixture was cooled to room temperature, dis-
solved in CH2Cl2, and poured onto an aluminum dish. Residual monomer
was removed in a vacuum drying cabinet at 608C for 12h. Conversion
was evaluated gravimetrically; molecular weight and polydispersity index
(PDI) were determined by size exclusion chromatography (SEC). Con-
version = 83%; Mn = 18600 gmolꢀ1; PDI = 1.12.
1,2-Bis(chlorodiisopropylsilanyl)benzene
(24):
1,2-Dibromobenzene
(0.36 g, 0.15 mmol) was added dropwise to magnesium turnings (1.66 g,
68.20 mmol) in THF (35 mL) in order to start the reaction. Diisopropyl-
chlorosilane (8.62g, 57.19 mmol) was added. 1,2-Dibromobenzene
(6.910 g, 29.29 mmol) in THF (50 mL) was added slowly to allow the re-
action to reflux gently before heating to reflux for 42h. After the mixture
was cooled to room temperature, hexane was added (110 mL). Filtration
of the salts, evaporation of the solvents in vacuo, and purification by FC
(pentane/Et3N
= 100:1) yielded the bis(diisopropylsilanyl)benzene
(1.34 g, 14%); 1H NMR (200 MHz): d = 7.53 7.49 (m, 2H, CH), 7.37
7.30 (m, 2H, CH), 4.30 (t, J = 3.5 Hz, 2H, SiH), 1.40 1.13 (m, 4H, CH),
1.12 0.91 ppm (m, 24H, CH3).
Bis(isopropylsilanyl)benzene (600 mg, 1.96 mmol) was saturated with
Acknowledgement
chlorine at 08C. Purging with argon removed excess chlorine and yielded
the product 24 (735 mg, >98%); H NMR (300 MHz): d = 7.87 (dd, J1
1
= 5.7 Hz, J2 = 3.4 Hz, 2H, CH), 7.41 (dd, J1 = 5.6 Hz, J2 = 3.4 Hz, 2H,
CH), 1.69 (sept., J = 7.3 Hz, 4H, CH), 1.14 (d, J = 7.1 Hz, 12H, CH3),
0.90 ppm (d, J = 7.3 Hz, 12H, CH3). The physical data are in agreement
with the values reported in literature.[15,16]
We thank the Deutsche Forschungsgemeinschaft (STU280/1-1) and the
Fonds der Chemischen Industrie (PhD stipend to C.W.) for funding. Dr.
Olaf Burghaus is acknowledged for assistance during the EPR experi-
ments as well as Manuel Ellermann for the preparation of alkoxyamine
15.
1,1,3,3-Tetraethyl-2-(1-phenyl-ethoxy)-2,3-dihydro-1H-benzo[1,2,5]azadi-
silole (25): To a solution of 1,2-bis(chlorodiethylsilyl)benzene (0.11 g,
0.35 mmol) and a catalytic amount of 4-(N,N-dimethylamino)pyridine in
dry DMF (2mL) in a sealed tube was added a mixture of O-(1-phenyle-
[1] a) D. H. Solomon, E. Rizzardo, P. Cacioli, US Patent 4581429; P.
Eur. Pat. Appl. 135280 (Chem. Abstr., 1985, 102, 221335q); b) M. K.
Georges, R. P. N. Veregin, P. M. Kazmaier, G. K. Hamer, Macromo-
lecules 1993, 26, 2987; c) Review: C. J. Hawker, A. W. Bosman, E.
Harth, Chem. Rev. 2001, 101, 3661.
thyl)hydroxylamine (32mg, 0.23 mmol) and NEt (71 mg, 0.70 mmol).
3
The reaction mixture was heated to 1008C for 8 min in a sealed tube
under microwave irradiation. The solvent was evaporated in vacuo, and
the crude product was dissolved in pentane. Filtration, evaporation of the
solvent in vacuo, and purification by FC (MTBE/pentane/NEt3
=
[2] a) K. Matyjaszewski, J. Xia, Chem. Rev. 2001, 101, 2921; b) M. Ka-
migaito, T. Ando, M. Sawamoto, Chem. Rev. 2001, 101, 3689.
[3] E. Rizzardo, J. Chiefari, R. T. A. Mayadunne, G. Moad, S. H. Thang,
In Controlled/Living Radical Polymerization (Ed.: K. Matyjaszew-
ski), American Chemical Society, Washington DC, 2000, ACS Sym-
posium Series 768, p. 278.
20:80:1) yielded the product 25 (69 mg, 78%). IR (film): n˜ = 3047s,
2874s, 1456m, 1411w, 1369w, 1233m, 1121m, 1075m, 1054m, 1006m,
929s, 876m, 760s, 700s cmꢀ1 1H NMR (300 MHz): d = 7.41 7.29 (m,
;
9H, CH), 4.58 (q, J = 6.6 Hz, 1H, CH), 1.47 (d, J = 6.6 Hz, 3H, CH3),
1.00 0.79 (m, 14H), 0.72ppm (t,
J
=
7.5 Hz, 6H, CH3); 13C NMR
(75 MHz): d = 144.36 (2C), 132.02 (2CH), 131.61 (2CH), 128.72 (CH),
128.41 (CH), 128.20 (CH), 127.71 (CH), 126.78 (CH), 126.33 (C), 84.7
[4] H. Fischer, Chem. Rev. 2001, 101, 3581.
[5] S. Marque, H. Fischer, E. Baier, A. Studer, J. Org. Chem. 2001, 66,
1146.
(CH), 23.04 (CH3), 7.31 (2CH ), 7.21 (2 CH3), 6.62(2CH 2), 6.20 ppm (2
3
CH2); MS (EI): 383 (14, [M]+), 279 (100), 207 (54), 179 (34), 105 (58);
[6] a) A. Goto, T. Terauchi, T. Fukuda, T. Miyamoto, Macromol. Rapid
Commun. 1997, 18, 673; b) T. Fukuda, A. Goto, Macromol. Rapid
Commun. 1997, 18, 683; c) W. G. Skene, S. T. Belt, T. J. Connolly, P.
Hahn, J. C. Scaiano, Macromolecules 1998, 31, 9103; d) K. Ohno, Y.
Tsujii, T. Miyamoto, T. Fukuda, M. Goto, K. Kobayashi, T. Akaike,
Macromolecules 1998, 31, 1064; e) S. A. F. Bon, G. Chambard, A. L.
German, Macromolecules 1999, 32, 8269; f) A. Goto, T. Fukuda,
Macromol. Chem. Phys. 2000, 201, 2138; g) D. Benoit, S. Grimaldi,
S. Robin, J.-P. Finet, P. Tordo, Y. Gnanou, J. Am. Chem. Soc. 2000,
122, 5929; h) S. Marque, C. LeMercier, P. Tordo, H. Fischer, Macro-
molecules 2000, 33, 4403; i) E. Drockenmuller, J.-M. Catala, Macro-
molecules 2002, 35, 2461; j) D. Bertin, F. Chauvin, S. Marque, P.
Tordo, Macromolecules 2002, 35, 3790; k) C. Le Mercier, S. Acerbis,
D. Bertin, F. Chauvin, D. Gigmes, O. Guerret, M. Lansalot, S.
Marque, F. Le Moigne, H. Fischer, P. Tordo, Macromol. Symp. 2002,
182, 225.
HRMS (EI): calcd for C22H33NOSi2: 383.2101; found: 383.2089.
1,1,3,3-Tetraisopropyl-2-(1-phenyl-ethoxy)2,3-dihydro-1H-benzo[1,2,5]-
azadisilole (26): To
a solution of 1,2-bis(chlorodiethylsilyl)benzene
(119 mg, 0.32mmol) and a catalytic amount of 4-( N,N-dimethylamino)-
pyridine in dry DMF (2mL) in a sealed tube was added a mixture of O-
(1-phenyl-ethyl)hydroxylamine (36 mg, 0.26 mmol) and NEt3 (80 mg,
0.79 mmol). The reaction mixture was heated to 908C for 10 min in a
sealed tube under microwave irradiation. The solvent was evaporated in
vacuo, and the crude product was dissolved in pentane. Filtration, evapo-
ration of the solvent in vacuo, and purification by FC (MTBE/pentane/
NEt3 = 20:80:1) yielded the product 26 (51 mg, 45%). IR (film): n˜ =
3048m, 2943s, 2891s, 1464s, 1383m, 1364m, 1115m, 1072m, 994m, 918s,
883s, 749m, 698s, 683s, 622m, 502m, 460m cm ꢀ1; 1H NMR (300 MHz): d
= 7.50 7.29 (m, 9H, CH), 4.61 (q, J = 6.5 Hz, 1H, CH), 1.50 (d, J =
6.6 Hz, 3H, CH3), 1.44 1.28 (m, 2CHSi), 1.27 1.11 (m, 2H, CH), 1.10
1.02ppm (m, 24H); 13C NMR (75 MHz): d = 144.3 (3C), 133.0 (2CH),
128.1 (2CH), 128.1 (2CH), 127.6 (CH), 126.9 (2CH), 84.2 (CH), 23.1
(CH3), 18.6 (CH3), 18.5 (CH3), 18.3 (CH3), 18.2(CH 3), 13.8 (2CH),
13.6 ppm (2CH); EI-MS: 439 (25, [ M]+), 367 (22), 365 (23), 335 (60),
333 (96), 331 (100), 277 (87), 269 (81); HRMS (EI) calcd for
C26H41NOSi2 ([M]+): 439.2727; found: 439.2724.
[7] T. Schulte, A. Studer, Macromolecules 2003, 36, 3078.
[8] G. Moad, E. Rizzardo, Macromolecules 1995, 28, 8722.
[9] S. Marque, J. Sobek, H. Fischer, A. Kramer, P. Nesvadba, W. Wun-
derlich, Macromolecules 2003, 36, 3440. For steric effects in five-
membered cyclic nitroxides, see: R. D. Puts, D. Y. Sogah, Macromo-
lecules 1996, 29, 3323.
Typical procedure for the polymerization of styrene: A Schlenk tube was
charged with the alkoxyamine initiator 13 (30 mg, 90.0 mmol) and styrene
(1.03 mL, 8.996 mmol) under argon. The styrene was previously degassed
in three freeze thaw cycles and sealed off under argon. The polymeriza-
tion was carried out under argon at 1058C for 24 h. The resulting mixture
was cooled to room temperature, dissolved in CH2Cl2, and poured onto
[10] G. B¸chi, H. W¸est, Helv. Chim. Acta 1979, 62, 2661.
[11] M. A. Blanchette, W. Choy, J. T. Davis, A. P. Essenfeld, S. Masa-
mune, W. R. Roush, T. Sakai, Tetrahedron Lett. 1984, 25, 2183.
[12] M. W. Rathke, M. Novak, J. Org. Chem. 1985, 50, 2624.
[13] W. R. Couet, R. C. Brasch, G. Sosnovsky, J. Lukszo, I. Prakash, C. T.
Gnewuch, T. N. Tozer, Tetrahedron 1985, 41, 1165.
1165
Chem. Eur. J. 2004, 10, 1156 1166
www.chemeurj.org
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

A. Studer et al.
FULL PAPER
[14] K. Matyjaszewski, B. E. Woodworth, X. Zhang, S. G. Gaynor, Z.
Metzner, Macromolecules 1998, 31, 5955.
[15] M. Ishikawa, H. Sakamoto, T. Tabuchi, Organometallics 1991, 10,
3173.
[16] a) A. Naka, T. Okada, M. Ishikawa, J. Organomet. Chem. 1996, 521,
163; b) D. S. Brown, P. T. Gallagher, A. P. Lightfoot, C. J. Moody,
A. M. Z. Slawin, E. Swann, Tetrahedron 1995, 51, 11473.
[17] a) D. Benoit, V. Chaplinski, R. Braslau, C. J. Hawker, J. Am. Chem.
Soc. 1999, 121, 3904; b) D. C. Tully, M. J. Roberts, B. H. Geierstang-
er, R. B. Grubbs, Macromolecules 2003, 36, 4302.
[18] S. Grimaldi, J.-P. Finet, A. Zeghdaoui, P. Tordo, D. Benoit, Y.
Gnanou, M. Fontanille, P. Nicol, J.-F. Pierson, PolymerPolym. Prep.
1997, 38, 651.
[19] Recently, Hintermann et al. reported the n-butyl acrylate polymeri-
zation with a cyclic nitroxide that gave rather good control (PDI =
1.2). However, only low molecular weight polybutyl acrylate was
prepared: T. Hintermann, A. Kramer, P. Nesvadba, J. Fink, Polymer-
Polym. Prep. 2002, 43, 86; C. A. Knoop, A. Studer, J. Am. Chem.
Soc. 2003, 125, 16327.
R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J. Och-
terski, G. A. Petersson, P. Y. Ayala, Q. Cui, K. Morokuma, D. K.
Malick, A. D. Rabuck, K. Raghvachari, J. B. Foresman, J. Cioslow-
ski, J. V. Oritz, A. G. Baboul, B. B. Stefanov, G. Liu, A. Liashenko,
P. Piskorz, I. Komaromi, R. Gomperts, R. L. Martin, D. J. Fox, T.
Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara, C. Gonzales,
M. Challacombe, P. M. W. Gill, B. G. Johnson, W. Chen, M. W.
Wong, J. L. Andres, M. Head-Gordon, E. S. Replogle, J. A. Pople,
Gaussian98, RevisionA.9 ed., 1998.
[22] F. B. van Duijneveldt, J. G. C. M. van Dunijeveldt-van de Rijdt, J. H.
van Lenthe, Chem. Rev. 1994, 94, 1873.
[23] a) G. A. Kovtun, A. L. Aleksandrov, V. A. Golubev, Bull. Acad. Sci.
USSR, Div. Chem. Sci. 1974, 10, 2115; b) J. A. Howard, J. C. Tait, J.
Org. Chem. 1978, 43, 4279; c) D. W. Grattan, D. J. Carlsson, J. A.
Howard, D. M. Wiles, Can. J. Chem. 1979, 57, 2834.
[24] a) A. Goto, Y. Kwak, C. Yoshikawa, Y. Tsujii, Y. Sugiura, T.
Fukuda, Macromolecules 2002, 35, 3520; b) S. Gennady, G. S. Anan-
chenko, H. Fischer, J. Polym. Sci. Part A: Polym. Chem. 2001, 39,
3604.
[20] A. W. Hui, A. E. Hamielec, J. Appl. Polym. Sci. 1972, 16, 749.
[21] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A.
Robb, V. G. Z. J. R. Cheeseman, J. A. Montgomery, Jnr., R. E. J.
Stratman, J. C. Burant, S. Dapprich, J. M. Millam, A. D. Daniels,
K. N. Kudin, M. C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi,
[25] Decomposition of 16 was not a clean reaction. In addition to the for-
mation of styrene, other unidentified decomposition products were
observed during the NMR studies.
Received: August 5, 2003
Revised: October 31, 2003 [F5427]
1166
¹ 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
Chem. Eur. J. 2004, 10, 1156 1166