Direct functionalization of labile alkoxyamines

Article abstract of DOI:10.1016/j.tetlet.2012.06.049

Direct esterification of a labile alkoxyamine R¹R ²NOR³, which was previously reported as unsuccessful, is achieved by a Mitsunobu reaction or a nucleophilic substitution. Ester derivatives are obtained under smooth conditions and easily purified. Macrocyclization attempts on ester derivatives were successful for five-membered ring lactones and unsuccessful for 13-membered ring lactones. Moreover, the success of the cyclization was dramatically dependent on the quality of the solution degassing. Poor degassing led to unexpected carbonate alkoxyamine.

Full text of DOI:10.1016/j.tetlet.2012.06.049

Tetrahedron Letters 53 (2012) 4543–4547
Contents lists available at SciVerse ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Direct functionalization of labile alkoxyamines
⇑
⇑
Paul Brémond , Kuanysh Kabytaev, Sylvain R. A. Marque
Université d’Aix-Marseille, Institut de Chimie Radicalaire—UMR 7273, case 551, Avenue Escadrille Normandie-Niemen, 13397 Marseille cedex 20, France
CNRS, Institut de Chimie Radicalaire—UMR 7273, case 551, Avenue Escadrille Normandie-Niemen, 13397 Marseille cedex 20, France
a r t i c l e i n f o
a b s t r a c t
Direct esterification of a labile alkoxyamine R1R2NOR3, which was previously reported as unsuccessful, is
achieved by a Mitsunobu reaction or a nucleophilic substitution. Ester derivatives are obtained under
smooth conditions and easily purified. Macrocyclization attempts on ester derivatives were successful
for five-membered ring lactones and unsuccessful for 13-membered ring lactones. Moreover, the success
of the cyclization was dramatically dependent on the quality of the solution degassing. Poor degassing led
to unexpected carbonate alkoxyamine.
Article history:
Received 22 March 2012
Revised 6 June 2012
Accepted 11 June 2012
Available online 20 June 2012
Keywords:
Alkoxyamine
Ó 2012 Elsevier Ltd. All rights reserved.
Esterification
Macrocyclization
Nitroxide
Persistent Radical Effect
Alkoxyamines R1R2NOR3 are labile molecules that can undergo
homolysis of the C–ON bond under moderate heating to release
nitroxides R1R2NOÅꢀand alkyl radicals R3Å Scheme 1). They are
widely used in Nitroxide Mediated Polymerization (NMP) as initi-
ator/controller agents,1,2 and in Radical Organic Chemistry as a
source of alkyl radicals.3 It is therefore interesting to obtain com-
plex alkoxyamines from functionalization of simple alkoxyamines:
for example, synthesis of esters or amides from alkoxyamines
bearing a carboxylic acid function.
Today, two of the most well known alkoxyamines are com-
pounds 1 and 2, based on the SG1 nitroxide (Scheme 1). However,
only limited functionalizations of alkoxyamines 1 and 2 have
been reported (Scheme 2). Esterification of 1 has been already
achieved,4,5 but attempts to esterify 2 have failed, although it has
successfully been converted into an amide such as 4.6
go condensation with a nucleophilic alcohol. The most widely used
method is the transformation of the carboxylic group into its cor-
responding acyl chloride by treatment with thionyl chloride. This
method was successfully applied on alkoxyamine 1, which carries
only one methyl group in the
a-position of the carboxylic group,
affording the corresponding acyl chloride.4,5,9 Similar attempts to
make the acyl chloride of alkoxyamine 2 were not successful.10
One should note that the carboxylic moiety of 2 is more hindered
than that of 1, as it contains an extra methyl group, and therefore
the carboxylic group is in neopentylic position. Another approach
relying on the use of DCC also failed.
During the last decade, relationship (1) was developed (at
T = 120 °C) to predict homolysis rate constant kd from the electrical
Hammett constant
r
r
I, the released radical stabilization constant
RS, and the steric effects of the alkyl fragment t 11
.
Then, assuming
Currently, if a specific ester derivative of alkoxyamine 2 is re-
quired, esterification has to be accomplished on the precursor of
the alkyl part of 2 before its coupling with the nitroxide. To allevi-
ate this difficulty, it would be desirable to have an efficient, direct
way to functionalize the labile alkoxyamine 2. Following our ongo-
ing program of synthesis and physico-chemical evaluation of new
alkoxyamines,7,8 we report herein the investigation of the direct
functionalization of the carboxylic moiety of alkoxyamine 2 into
different esters.
R1
R2
R3
R1
R2
kd
kc
N
O
N
O
R3
O
O
N
O
OH
N
O
OH
N
O
Esterification reactions are generally achieved by activating a
carboxylic function to a more electrophilic species that can under-
P(O)(OEt)2
P(O)(OEt)2
P(O)(OEt)2
Ea = 132.0 kJ/mol
Ea = 112.3 kJ/mol
1
2
SG1
⇑
Corresponding authors.
E-mail addresses: paul.bremond@univ-amu.fr (P. Brémond), sylvain.marque@
Scheme 1. Homolysis of alkoxyamines and structures of labile alkoxyamines 1 and 2
univ-amu.fr (S.R.A. Marque).
(Ea: activation energy).
0040-4039/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.tetlet.2012.06.049

4544
P. Brémond et al. / Tetrahedron Letters 53 (2012) 4543–4547
O
O
O
Recently, we showed a striking increase in kd upon protonation
or quaternization of the alkyl fragment, that is, increase in the elec-
tron withdrawing capacities of the alkyl fragment supporting the
increase of kd upon activation of the carboxylic group.7,8 However,
as N-hydroxysuccinimide (NHS) is a good nucleophile, it reacted
fast enough with the DCC intermediate to afford the second-hand
alkoxyamines suitable for reaction with amines and then affords
amide 4 (Scheme 2).6
Analysis of relationship (1) in conjunction with the previous
experimental results made it clear that esterification of 2 via acti-
vation of the carboxylic acid was likely to fail, with the possible
exception of reaction with a highly nucleophilic alcohol. On the
other hand, if the acid was deprotonated to form its carboxylate
1) SOCl2
2) ROH
yield: ca. 50 %
N
O
OH
N
O
R
P(O)(OEt)2
P(O)(OEt)2
1
3
O
O
1) DCC
NHS
N
O
OH
N
O
HN
R
R
yield: 70 %
2) RNH2
P(O)(OEt)2
P(O)(OEt)2
2
4
DCC , ROH
or
O
O
O
(rI,COOꢁ = ꢁ0.13), the C–ON bond homolysis would be deacti-
vated,16 and hence 2 might be derivatized into its ester under mild
conditions. As a result, we changed our esterification strategy from
activating the carboxylic acid moiety making the alcohol compo-
nent to be the electrophilic species. First, we thought about the
Mitsunobu reaction to save the pre-activation step of alcohol. After
screening several conditions, the best result was obtained by add-
ing alkoxyamine 2 and then 10-undecenol to a mixture of triphen-
ylphosphine and diisopropylazodicarboxylate (DIAD) in THF.
Obtaining the targeted ester 6 by direct functionalization of 2
was gratifying, despite a low 11% yield (Scheme 3a).17 However,
the Mitsunobu reaction is not suitable for large-scale reactions,
due to the excess of reagents (Ph3P, DIAD, acid and/or alcohol)
and the byproducts obtained (Ph3P@O, reduced DIAD), the latter
requiring time-consuming and expensive purification techniques.
It is also known that the Mitsunobu reaction can proceed through
several different mechanistic pathways involving either activation
of the alcohol, or activation of the acid.18 These facts might explain
partly why low yields were obtained.
N
O
OH
N
O
yield: < 5 %
or decomposition
1) SOCl2
2) ROH
P(O)(OEt)2
P(O)(OEt)2
2
5
Scheme 2. Functionalization of alkoxyamines
amidification, respectively.
1 and 2 by esterification and
that the effect of the acyl chloride COCl, the DCC-activated carbox-
ylic group (CODCC) and the carboxylic group COOH functions are
very similar on the stabilization of the released radical
(
r
RS = 0.21)11 and the bulkiness (
t
= 1.24)11 of the alkyl fragment,
COOH = 0.36)12
COCl = 0.50)12 or DCC-activated13 carboxylic
the polar effect when going from carboxylic group (
to acyl chloride (
r
r
group leads to a decrease in Ea of roughly 5 kJ/mol (Ea = 107.0 kJ/
mol) to the COCl function, that is, a 4-fold increase in kd at room
temperature.14 This increase in the dissociation constant in combi-
nation with a low reactivity of both the alcohol (poor nucleophile)
and the carboxylic group in the neopentylic-like position favors the
decomposition of intermediate acyl chloride15 rather than the for-
mation of the expected ester 5.
We then turned our attention to an even simpler method, using
pre-activated alcohol such as halides or mesylates. After deproto-
nation of 2 with 1,8-diazabicyclo[5.4.0]undec-7-ene (DBU) in ben-
zene, the resulting carboxylate was reacted with methyl iodide at
room temperature for 24 h. The corresponding methyl ester 7b
was obtained in good yield (77%) after simple short filtration on sil-
ica gel, without work-up of the reaction.19,20 The same conditions
were successfully applied to ethyl iodide or allyl bromide and
afforded ethyl ester 8b in 78% yield21 and allyl ester 9b in 57%
yield,22 respectively. No traces of SG1 were detected during these
experiments, indicating that the homolysis of alkoxyamine species
did not occur (Scheme 3b).
Compound 2 is one of the most efficient alkoxyamines for
NMP.1 Along its invaluable kinetic properties for NMP, the pres-
ence of a carboxylic group is very appealing for pre- or postpoly-
merization transformations. Such versatility has been nicely
highlighted by the preparation of copolymers and armpolymers.23
Thus, we decided to test our procedure of esterification by grafting
a polyethylene glycol onto 2 to obtain 10b (Scheme 3c). First, we
tested the DBU/benzene procedure using monomethyl polyethyl-
eneglycol mesylate 10a as the electrophile, but no new products
were observed by 31P NMR. After screening several solvents (ben-
zene, CH2Cl2, THF, DMF), smooth conditions using Cs2CO3 as the
base and DMF as the solvent (24 h, room temperature) afforded
the targeted grafted polymer 10b in 40% conversion by 31P NMR.
Polymer was characterized by 31P NMR and grafting was confirmed
by ESI/MS.24,25
logðkd=sꢁ1Þ ¼ ꢁ14:3 þ 15:3 ꢂ rRS þ 19:5 ꢂ rI þ 7:0 ꢂ
m
ð1Þ
a
b
O
O
O
10-undecenol
N
O
OH
N
O
PPh3 , DIAD
THF, 0 ºC
8
P(O)(OEt)2
P(O)(OEt)2
2
6
yield: 11 %
1) DBU
Benzene
r.t. , 1 h
O
O
N
O
OH
N
O
O
R
2) R
X
P(O)(OEt)2
P(O)(OEt)2
Benzene
r.t. , 24 h
2
yield:
77 %
78 %
57 %
7a R X = Me
8a R X = Et
9a R X =
I
7b R = Me
8b R = Et
9b R =
I
Br
1) Cs2CO3
DMF
r.t. , 1 h
c
O
O
As exemplified with 1, alkoxyamines carrying an alkenyl frag-
ment are prone to undergo radical cyclization under thermal con-
ditions (T = 110 °C).4 A similar behavior for ester derivatives of
alkoxyamine 2 was expected but at lower temperatures. Macrocyc-
lization should occur by heating derivative 6 at 70 °C in t-BuOH26
and should afford either 12-endo-trig cyclic product 11 and/or
11-exo-trig product 12. Full conversion (48 h, T = 70 °C, degassing
N
O
OH
N
O
O
O
n
OMs
n
10a
2)
O
P(O)(OEt)2
P(O)(OEt)2
DMF
r.t. , 24 h
2
yield: 40 %
10b
Scheme 3. Direct esterification of alkoxyamine 2 by (a) Mitsunobu reaction, (b)
nucleophilic substitution of alkyl halides, (c) application to polymer grafting.

P. Brémond et al. / Tetrahedron Letters 53 (2012) 4543–4547
4545
O
O
O
O
O
70 ºC
N
O
N
O
O
N
O
(EtO)2(O)P
or
t-BuOH
8
P(O)(OEt)2
P(O)(OEt)2
6
11
12
12-endo-trig
11-exo-trig
70 ºC
t-BuOH
O
O
O
8
O
N
OH
P(O)(OEt)2
14
N O
+
+
8
P(O)(OEt)2
13
15
δ P = 24.8 ppm
O
H P
OEt
OEt
N
O
+
δ P = 7.1 ppm
16
Scheme 4. Attempt of macrocyclization on alkoxyamine 6.
O
O
O
+
SG1
70 ºC
OR
SG1
COOR
COOR
O2
N
O
OR
N
t-BuOH
P(O)(OEt)2
P(O)(OEt)2
6
R = 10-undecenyl
15 R = 10-undecenyl
18 R = Et
8b R = Et
OO
COOR
9b R = Allyl
19 R = Allyl
Scheme 6. Structure of compounds 15, 18, and 19.
typical of alkene 13 was observed by 1H NMR (d = 6.10 ppm) con-
firming the occurrence of the H-atom abstraction reaction. How-
ever, the presence of other vinylic protons discarded the
presence of the expected cyclic compounds 11 and/or 12. After
purification of the compound corresponding to the peak at
d = 24.8 ppm in 31P NMR, 1H NMR confirmed the presence of
vinylic protons precluding the cyclic compounds 11 and/or 12
and showed the absence of two methyl groups. ESI–MS analysis
showed a loss of 42 units of mass and 13C NMR showed that indeed
two methyl groups and a quaternary carbon were missing, when
comparing with the starting materials. The same observations
were made for 8b and 9b for samples prepared in the same way.
Furthermore, when experiments were performed in deuterated
benzene in an NMR probe, the typical signal of acetone was
recorded. Consequently, it led us to assume that the reactivity
observed can be rationalized invoking the conventional decompo-
sition of the tetroxide 17 as intermediate (Scheme 5). That is, as
the cross-coupling between nitroxide and tertiary alkyl radical
was strongly disfavored, the alkyl radical reacted with oxygen to
afford an alkyl peroxyl radical. As the latter had no other possibil-
ities than to react by self-termination, it afforded a tetroxide
species 17 that spontaneously decomposed into oxygen and
alkoxyl radicals. As alkoxyl radicals cannot react with nitroxide
and as no labile H-atoms were available, its b-fragmentation was
favored to afford acetone and the stabilized acyl radical.30,31 Then,
the latter reacted by cross-coupling to afford by-products 15, 18,
and 19 from 6, 8b, and 9b, respectively (Scheme 6).32–34
17
ROOC
O
O
O
O
COOR
O2
H
ROOC
O
-
abstraction
scission
-
β
O
ROOC
OH
+
ROOC
SG1
ROOC
O
O
COOR
SG1 COOR
Scheme 5. Proposed mechanism for the formation of the side-product SG1-COOR.
under moderate vacuum, P = 10ꢁ2 Torr) was observed by 31P NMR,
and only two peaks were recorded at d = 24.8 ppm (80%) and
d = 7.1 ppm (20%), the latter corresponding to diethyl phosphite
(EtO)2P(O)H. Indeed, H-atom abstraction from the alkyl radical
by the nitroxide is well documented,27 and in our case, it led to
(undec-10-enyl)methacrylate 13 and hydroxylamine 14 (Scheme
4). The latter is reported to be thermally unstable and decomposes
into nitrone 16 and diethylphosphite.28,29 The presence of a signal

4546
P. Brémond et al. / Tetrahedron Letters 53 (2012) 4543–4547
Acknowledgments
O
O
O
O
70 ºC
O
+
O
N
O
N
O
The authors thank Aix-Marseille University and CNRS for
financial support. K.K. greatly thanks the Agence Nationale de la
Recherche for a post-doctoral Grant (ANR NITROMRI ANR-09-
BLAN-0017-01). We thank Prof. L. Charles for fruitful discussions
on ESI–MS data.
t-BuOH
P(O)(OEt)2
P(O)(OEt)2
O
H P
OEt
OEt
+
+
9b
20a,b
by-product
References and notes
Scheme 7. Cyclization of compound 9b.
1. Gigmes, D.; Marque, S. In Nitroxide-Mediated Polymerization and its Applications
in Encyclopedia of Radicals in Chemistry Biology and Materials; Chatgilialoglu, C.,
Studer, A., Eds.; John Wiley & Sons Ltd: Chichester, UK, 2012; pp 1813–1850.
2. Bertin, D.; Gigmes, D.; Marque, S. R. A.; Tordo, P. Chem. Soc. Rev. 2011, 40, 2189–
2198.
3. Tebben, L.; Studer, A. Angew. Chem., Int. Ed. 2011, 50, 5034–5068.
4. Bertin, D.; Gigmes, D.; Marque, S. R. A.; Tordo, P. Tetrahedron 2005, 61, 8752–
8761.
5. Bertin, D.; Gigmes, D.; Marque, S. R. A.; Maurin, R.; Tordo, P. J. Polym. Sci., Part A:
Polym. Chem. 2004, 42, 3504–3515.
6. Gigmes, D.; Vinas, J.; Chagneux, N.; Lefay, C.; Phan, T. N. T.; Trimaille, T.; Dufils,
P.-E.; Guillaneuf, Y.; Carrot, G.; Boué, F.; Bertin, D. ACS Symp. 2009, 1024, 245–
262.
When experiments were performed on samples degassed under
high vacuum (48 h, T = 70 °C, degassing under high vacuum, P = 10
ꢁ5 Torr) a totally reverted reactivity was observed, that is, 20% of
side-products 15, 18, 19 and 80% of diethylphosphite, and new
trace peaks were also observed for 9b (vide infra).35 It nicely con-
firmed the influence of oxygen on the reactivity and its role in
the proposed mechanism. The kinetics and the consequence of this
mechanism on the reactivity of labile alkoxyamines are currently
under investigation.29 For 9b experiment was again performed in
high vacuum-sealed tube and peaks at d = 23.0 and d = 23.3 ppm
were observed by 31P NMR along with 14 and diethylphosphite.
It is likely that the formation of large rings (14 to 15 members)
is strongly entropically disfavored and cannot compete with the
H-atom abstraction leading to hydroxylamine and alkyl methacry-
lates involving that large rings cannot be prepared with this proce-
dure. The formation of five-membered ring lactone was reported
for alkoxyamine based on 1.4 The new peaks aforementioned are
in the expected zone for this family of compounds. Thus, experi-
ment was performed on a preparative scale by heating a t-BuOH
solution of 9 degassed under high vacuum (P = 10ꢁ5 Torr) in a
sealed Schlenk at 120 °C for 2 h. Full conversion was reached by
31P NMR, affording 66% of 20a and 20b (50% isolated yield) and
34% of diethylphosphite. The vinylic protons at d = 5.89 ppm con-
firmed the presence of allyl methacrylate 21 and hence the mech-
anism of its generation (Schemes 4 and 7). The 1H, 31P, and 13C
NMR confirmed the 5-exo-trig-cyclization as no vinylic protons
were detected, as new CH2O (d = 71.78 ppm and d = 67.80 ppm
for 20a and d = 72.13 ppm and d = 67.99 ppm for 20b) and CH
(d = 43.18 ppm for 20a and d = 42.92 ppm for 20b) signals were
arising.36 Furthermore the HRMS confirmed the expected mass.37
In conclusion, we achieved direct esterification of the labile alk-
oxyamine 2 by two different methods: (1) a Mitsunobu reaction
followed by tedious purification, (2) a very simple nucleophilic dis-
placement of an alkyl halide by the carboxylate anion of 2 followed
by simple filtration through silica gel. This last method allows sev-
eral ester derivatives of the labile alkoxyamine 2 to be obtained di-
rectly and efficiently without pre-coupling functionalization. The
preparation of the polymeric initiator 10b in good yield highlights
the potential of such type of functionalization for applications in
Polymer Science.
7. Brémond, P.; Marque, S. R. A. Chem. Commun. 2011, 47, 4291–4293.
8. Brémond, P.; Koïta, A.; Marque, S. R. A.; Pesce, V.; Roubaud, V.; Siri, D. Org. Lett.
2012, 14, 358–361.
9. Bertin, D.; Gigmes, D.; Marque, S. R. A.; Milardo, S.; Peri, J.; Tordo, P. Collect.
Czech. Chem. Commun. 2004, 69, 2223–2238.
10. Experimental conditions have not been reported in Ref. 6. However conditions
are expected to be the same than in Ref. 4, that is addition of 3 equiv of SOCl2 to
a solution of the acid in CH2Cl2 followed by stirring at room temperature.
11. Bertin, D.; Gigmes, D.; Marque, S. R. A.; Tordo, P. Macromolecules 2005, 38,
2638–2650.
12. Charton, M. Prog. Phys. Org. Chem. 1981, 13, 119–251.
13. The Hammett values for the intermediate COOC(@NR)NR cannot be
straightforwardly estimated. However, the F value of OC(@NR) is expected to
be larger than of OH (FOH = 0.33) due to electron withdrawing atom and lower
than OCN group (FOCN = 0.69), meaning that the effect should be closed to the
effect of Cl (FCl = 0.42) and hence of COCl. See: Hansch, C.; Leo, A.; Taft, R. W.
Chem. Rev. 1991, 91, 165–195.
14.
rCMe2COOH = 0.07, r
CMe2COCl = 0.13, and A = 2.4 1014 sꢁ1
15. Preparation of pivaloyl chloride by reaction of pivalic acid with SOCl2 requires
warming of the solution above 50 °C (see for example Attygalle, A. B.;
Nishshanka, U.; Weisbecker, C. S. J. Am. Soc. Mass. Spectrom. 2011, 1515–
1525. As warming is precluded with labile alkoxyamine such as 2, it may occur
that the formation of pivaloyl-type chloride was very slow at room
temperature, favoring the occurrence of side-reactions and hence the
decomposition of the starting material..
16. Ea cannot be estimated with high reliability due to intimate ion pair effect
which depends dramatically on the solvent, see: Audran, G.; Brémond, P.;
Marque, S. R. A.; Obame, G., submitted for publication . However, the homolysis
will not be faster than for 2, see Bertin, D.; Gigmes, D.; Marque, S. R. A.; Siri, D.;
Tordo, P.; Trappo, G. Chem. Phys. Chem. 2008, 9, 272–281.
17. Undec-10-en-1-yl 2-((tert-butyl(1-(diethoxyphosphoryl)-2,2-dimethyl-propyl)-
amino)oxy)-2-methylpropanoate (6). 1H NMR (300 MHz, CDCl3): d 1.09 (9H, s),
1.19 (9H, s), 1.25–1.32 (18H, m), 1.59 (3H, s), 1.70 (3H, s), 3.27 (1H, d), 3.90–
4.10 (4H, m), 4.32–4.38 (2H, m), 4.92–5.01 (2H, m), 5.77–5.83 (1H, m) ppm. 13
C
NMR (75 MHz, CDCl3): d 16.18, 16.27, 16.57, 16.65, 25.95, 28.13, 28.24, 28.52,
28.86, 29.04, 29.18, 29.35, 29.42, 30.01, 30.08, 33.75, 35.93, 36.01, 61.74, 61.82,
62.17, 64.95, 69.19, 71.02, 83.84, 114.12, 139.10, 175.29 ppm. 31P NMR
(121 MHz, CDCl3):
d 25.3 ppm. HRMS (ESI): m/z calcd for C28H57N1O6P1
(M+H)+ 534.3918, found 534.3917.
18. But, T. Y. S.; Toy, P. H. Chem. Asian J. 2007, 2, 1340–1355.
19. General procedure for the synthesis of ester derivatives 5: 1,8-
diazabicyclo[5.4.0]undec-7-ene (DBU) (1.0 equiv) was added dropwise to a
stirred solution of alkoxylamine 2 (1.0 equiv) in dry benzene. The mixture was
stirred for 30 min under argon, and then the electrophile (1.0 equiv) was
added. After 24 h of stirring at room temperature, the solvent was removed in
vacuo and the crude material was purified by column chromatography on silica
gel (diethylether/hexanes) to give pure ester (yields: 7b: 77%; 8b: 78%; 9b:
57%).
We showed that the radical cyclization of large rings cannot be
performed with alkoxyamines derived from 2. On the other hand,
formation of five-membered ring lactone was performed in moder-
ate yield. In fact, with these examples, the limits of the Persistent
Radical Effect are reached, as the cross-coupling reaction is so slow
that it cannot compete with the H-atom abstraction side-reaction
and cannot play its role of prolonging the lifetime of the alkoxy-
amine until it reacts through the targeted reaction.
We also showed that the presence of oxygen might be an
important issue as it led to new by-products. This reaction took
place only because the cross-coupling reaction between nitroxide
and alkyl radical was extremely slow. The consequences of this ki-
netic behavior on Nitroxide Mediated Polymerization issue are
currently under investigation and will be reported in due course.
20. Beaudoin, E.; Bertin, D.; Gigmes, D.; Marque, S. R. A.; Siri, D.; Tordo, P. Eur. J.
Org. Chem. 2006, 1755–1768.
21. Ethyl 2-((tert-butyl(1-(diethoxyphosphoryl)-2,2-dimethylpropyl)amino)oxy)-2-
methylpropanoate (8b). 1H NMR (300 MHz, CDCl3): d 1.10 (9H, s), 1.19 (9H, s),
1.26–1.32 (9H, m), 1.59 (3H, s), 1.70 (3H, s), 3.27 (1 H, d, J = 26.0 Hz), 3.91–4.03
(1H, m), 4.07–4.11 (1H, m), 4.12–4.19 (2H, m), 4.33–4.39 (2H, m) ppm. 13C
NMR (75 MHz, CDCl3): d 14.08, 16.21, 16.30, 16.56, 16.64, 22.45, 28.13, 29.98,
30.05, 35.95, 36.04, 60.77, 61.77, 61.85, 62.17, 69.24, 71.05, 83.74, 175.26 ppm.
31P NMR (121 MHz, CDCl3):
C
d 25.3 ppm. HRMS (ESI): m/z calcd for
19H41N1O6P1 (M+H)+ 410.2666, found 410.2673.
22. Allyl 2-((tert-butyl(1-(diethoxyphosphoryl)-2,2-dimethylpropyl)amino)oxy)-2-
methylpropanoate (9b). 1H NMR (300 MHz, CDCl3): d 1.10 (9H, s), 1.19 (9H, s),

P. Brémond et al. / Tetrahedron Letters 53 (2012) 4543–4547
4547
1.28 (3H, t, J = 7.0 Hz), 1.29 (3H, t, J = 7 Hz), 1.62 (3H, s), 1.71 (3H, s), 3.27 (1H,
d, J = 26.0 Hz), 3.91–3.97 (1H, m), 4.03–4.09 (1H, m), 4.30–4.37 (2H, m), 4.59
(2H, d, J = 6 Hz), 5.23–5.34 (2H, m) ppm. 13C NMR (75 MHz, CDCl3): d 16.21,
16.30, 16.56, 16.64, 22.62, 27.99, 29.99, 30.06, 35.96, 36.04, 61.86, 61.94, 62.24,
65.53, 69.25, 71.06, 83.77, 118.68, 132.11, 174.81 ppm. 31P NMR (121 MHz,
CDCl3): d 25.3 ppm. HRMS (ESI): m/z calcd for C20H41N1O6P1 (M+H)+ 422.2666,
found 422.2669.
33. Diethyl (1-(tert-butyl((ethoxycarbonyl)oxy)amino)-2,2-dimethylpropyl)phospho-
nate (18). 1H NMR (300 MHz, CDCl3): d 1.13 (9H, s), 1.20 (9H, s), 1.26–1.34
(9H, m), 3.26 (1H, d, J = 24.0 Hz), 4.06–4.27 (6H, m) ppm. 13C NMR (75 MHz,
CDCl3): d 14.29, 16.28, 16.34, 16.50, 16.56, 27.57, 28.34, 28.40, 36.13, 36.20,
60.14, 61.87, 62.72, 64.20, 67.82, 69.14, 155.94 ppm. 31P NMR (121 MHz,
CDCl3): d 24.7 ppm. HRMS (ESI): m/z calcd for C16H35N1O6P1 (M+H)+ 368.2197,
found 368.2198.
23. Dire, C.; Nicolas, J.; Brusseau, S.; Charleux, B.; Magnet, S.; Couvreur, L. ACS
Symp. 2009, 1024, 303–318.
24. The polymer 10a (MW = 2000 g/mol) was purchased from Sigma–Aldrich and
34. Diethyl
(1-((((allyloxy)carbonyl)oxy)(tert-butyl)amino)-2,2-dimethyl-propyl)
phosphonate (19). 1H NMR (300 MHz, CDCl3): d 1.14 (9H, s), 1.15 (9H, s),
1.25–1.28 (6H, m), 3.22 (1H, d, J = 24.0 Hz), 3.90–4.05 (4H, m), 5.09–5.27 (2H,
m), 5.92–6.03 (1H, m) ppm. 13C NMR (75 MHz, CDCl3): d 16.21, 16.29, 16.52,
16.59, 27.67, 29.69, 29.76, 35.54, 35.42, 58.86, 58.97, 61.74, 77.22, 78.26,
117.58, 134.03 ppm. 31P NMR (121 MHz, CDCl3): d 24.5 ppm. HRMS (ESI): m/z
calcd for C17H34N1O6P1 (M+H)+ 380.2197, found 380.2195.
used as received. The grafted polymer 10b exhibits
25.1 ppm on 31P NMR, which is within the range of the chemical shifts of ester
derivatives of 2. The ESI–MS spectrum of 10b shows distribution of
a chemical shift of
a
polyethyleneglycol oligomers adducted with cesium for which the sum of
the end-group masses (396 Da) was found to be consistent with the expected
terminations within a 1.6% error.
35. Some oxidation products were still observed because the vacuum cap-closed
tube is expected to be less tight all along the experimental period than the
vacuum-sealed glass tube and leaks might occur.
36. The formation of six-membered ring (6-endo-trig-cyclization) is forbidden by
the Baldwin’s rules, and consequently such cyclization is not fast enough to
compete with the side-reaction aforementioned.
37. Diethyl (1-(tert-butyl((4,4-dimethyl-5-oxotetrahydrofuran-3-yl) methoxy)amino)-
2,2-dimethylpropyl)phosphonate (20a,b 1:1 mixture of diastereomers). 1H NMR
(300 MHz, CDCl3): d 1.08 (3H, s), 1.10 (3H, s), 1.13 (9H, s), 1.14 (9H, s), 1.15 (9H,
s), 1.16 (9H, s), 1.26–1.33 (14H, m), 3.23 (1H, d, J = 24.9 Hz), 3.25 (1H, d,
J = 25.1 Hz), 3.89–4.19 (14H, m), 4.40 (1H, dd, J = 7.2; 9.0 Hz), 4.47 (1H, dd,
J = 7.6; 9.5 Hz) ppm. 13C NMR (75 MHz, CDCl3): d 14.60, 14.83, 16.85, 17.40,
21.77, 22.28, 24.47. 25.81, 25.90, 26.07, 26.60, 26.68, 27.92. 28.42, 28.65, 33.51,
38.80, 40.76, 42.92, 43.18, 57.83, 59.21, 60.10, 60.16, 66.03, 67.06, 67.24, 67.80,
71.78, 72.13, 177.74, 180.07 ppm. 31P NMR (121 MHz, C6D6): d 23.0, 23.2 ppm.
HRMS (ESI): m/z calcd for C20H40N1O6P1 (M+H)+ 422.2666, found 422.2666.
25. At longer reaction time, degradation products from 2 and 10b were observed,
without improving the yield of the grafting.
26. Solution was degassed by 3 freeze-thaw cycles (P = 10ꢁ2 Torr).
27. Ananchenko, G. S.; Fischer, H. J. Polym. Sci., A: Polym. Chem. 2001, 39, 3604–
3621.
28. Bertin, D.; Gigmes, D.; Marque, S. R. A.; Tordo, P. e-Polymers 2003, 2, 1–9.
29. Bagryanskaya, E.; Edeleva, M.; Kabytaev, K.; Marque, S. R. A. Personnal
communication.
30. Ingold, K. U. Acc. Chem. Res. 1969, 2, 1–9.
31. Howard, J. A.; Ingold, K. U. J. Am. Chem. Soc. 1968, 90, 1056–1058.
32. Diethyl (1-(tert-butyl(((undec-10-en-1-yloxy)carbonyl)oxy)amino)-2,2-dimethyl
propyl)phosphonate (15). 1H NMR (300 MHz, CDCl3): d 1.13 (9H, s), 1.20 (9H,
s), 1.26–1.36 (18H, m), 1.57–1.68 (2H, m), 2.0–2.07 (2H, m), 3.26 (1H, d,
J = 24.0 Hz), 4.05–4.30 (6H, m), 4.91–5.01 (2H, m), 5.74–5.88 (1H, m) ppm. 13
C
NMR (75 MHz, CDCl3): d 16.28, 16.37, 16.50, 16.57, 27.59, 28.33, 28.40, 28.67,
28.90, 29.07, 29.35, 29.41, 33.87, 58.59, 58.88, 61.74, 61.82, 62.17, 64.95, 69.19,
71.02, 114.10, 139.18, 156.06 ppm. 31P NMR (121 MHz, CDCl3): d 24.8 ppm.
HRMS (ESI): m/z calcd for C25H51N1O6P1 (M+H)+ 492.3449, found 492.3442.